Merge remote-tracking branch 'origin/main' into sweep/2644

# Conflicts:
#	tests/process-global/cleanLinuxHopperRuntime.test.ts

.github/workflows/ci.yml

@@ -741,6 +741,7 @@ jobs:
           tests/boundary/filesystem/javascriptApplicationWorkflows.test.ts
           tests/boundary/filesystem/webScriptManifestPaths.test.ts
           tests/boundary/process/ProcessEvidence.host.test.ts
+          tests/boundary/process/windowsOutputCapture.test.ts
           tests/acceptance/webModuleCli.test.ts
           tests/acceptance/webSourceLocationCli.test.ts
           tests/boundary/cli/referenceSourceImportStatus.test.ts

README.md

@@ -344,7 +344,7 @@ Report vulnerabilities through [SECURITY.md](SECURITY.md).
 
 ## Star history
 
-🎉 **200,000 GitHub stars — thank you!**
+🎉 **250,000 GitHub stars — thank you!**
 
 Thanks to everyone using REA, reporting bugs, requesting features, testing builds, and contributing fixes.
 

bridge/ghidra/ReaGhidraBridge.java

@@ -934,7 +934,7 @@ public final class ReaGhidraBridge extends HeadlessScript {
             String value = decompiledText(decompile(function));
             JsonObject result = functionObservation(
                 value == null ? JsonNull.INSTANCE : GSON.toJsonTree(value),
-                noReturnCallLimitations(
+                functionBoundaryLimitations(
                     function,
                     currentProgram.getListing().getInstructions(function.getBody(), true)
                 )
@@ -1027,7 +1027,7 @@ public final class ReaGhidraBridge extends HeadlessScript {
         InstructionScan scan = scanInstructions(function, Integer.MAX_VALUE);
         return functionObservation(
             GSON.toJsonTree(renderAssembly(scan.instructions)),
-            noReturnCallLimitations(function, scan.instructions)
+            functionBoundaryLimitations(function, scan.instructions)
         );
     }
 
@@ -1046,7 +1046,7 @@ public final class ReaGhidraBridge extends HeadlessScript {
         limitations.add(
             "The fast path does not invoke the decompiler or scan whole-program names and strings."
         );
-        List<String> boundaryLimitations = noReturnCallLimitations(function, scan.instructions);
+        List<String> boundaryLimitations = functionBoundaryLimitations(function, scan.instructions);
         for (String limitation : boundaryLimitations) limitations.add(limitation);
         JsonObject result = new JsonObject();
         result.add("procedure", procedureIdentity(function));
@@ -1069,7 +1069,7 @@ public final class ReaGhidraBridge extends HeadlessScript {
             result.add(canonicalAddress(item.getEntryPoint()));
         }
         if (callers) {
-            JsonObject observation = functionObservation(result, noReturnCallLimitations(function,
+            JsonObject observation = functionObservation(result, functionBoundaryLimitations(function,
                 currentProgram.getListing().getInstructions(function.getBody(), true)
             ));
             observation.add("relationship_coverage", callRelationshipCoverage(null,
@@ -1099,7 +1099,7 @@ public final class ReaGhidraBridge extends HeadlessScript {
         result.add("classification", functionClassification(function));
         return functionObservation(
             result,
-            noReturnCallLimitations(
+            functionBoundaryLimitations(
                 function,
                 currentProgram.getListing().getInstructions(function.getBody(), true)
             )
@@ -1131,12 +1131,22 @@ public final class ReaGhidraBridge extends HeadlessScript {
         result.addProperty("direction", direction);
         result.add("references", edges);
         result.addProperty("reference_kinds_available", true);
-        JsonArray unresolvedCalls = direction.equals("outgoing")
-            ? callBoundaryObservation(
+        List<String> limitations;
+        JsonArray unresolvedCalls = new JsonArray();
+        if (direction.equals("outgoing")) {
+            CallBoundaryObservation calls = callBoundaryObservation(
                 function, currentProgram.getListing().getInstructions(function.getBody(), true), null, true, references
-            ).unresolvedCalls() : new JsonArray();
+            );
+            unresolvedCalls = calls.unresolvedCalls();
+            limitations = calls.limitations();
+        } else {
+            limitations = functionBoundaryLimitations(
+                function,
+                currentProgram.getListing().getInstructions(function.getBody(), true)
+            );
+        }
         result.add("unresolved_calls", unresolvedCalls);
-        return result;
+        return functionObservation(result, limitations);
     }
 
     private JsonObject inspectNativeDataType(JsonObject params) throws Exception {
@@ -1337,11 +1347,7 @@ public final class ReaGhidraBridge extends HeadlessScript {
         result.add("references", references);
         if (includeIncoming) {
             TreeMap<String, Reference> incoming = new TreeMap<>();
-            ReferenceIterator incomingIterator = currentProgram.getReferenceManager().getReferencesTo(address);
-            while (incomingIterator.hasNext()) {
-                monitor.checkCancelled();
-                addReference(incoming, incomingIterator.next());
-            }
+            visitObservedReferencesTo(address, reference -> addReference(incoming, reference));
             List<Reference> orderedIncoming = new ArrayList<>(incoming.values());
             orderedIncoming.sort(REFERENCE_ORDER);
             JsonArray incomingReferences = new JsonArray();
@@ -1393,16 +1399,8 @@ public final class ReaGhidraBridge extends HeadlessScript {
         requireKeys(params, Set.of("document", "address"));
         requireDocument(params);
         Address address = requireAddress(params, "address");
-        ReferenceIterator references = currentProgram.getReferenceManager().getReferencesTo(address);
         Set<Address> sources = new HashSet<>();
-        while (references.hasNext()) {
-            monitor.checkCancelled();
-            Reference reference = references.next();
-            if (reference.isEntryPointReference()) {
-                continue;
-            }
-            sources.add(reference.getFromAddress());
-            }
+        visitObservedReferencesTo(address, reference -> sources.add(reference.getFromAddress()));
         List<Address> ordered = new ArrayList<>(sources);
         ordered.sort(Address::compareTo);
         JsonArray result = new JsonArray();
@@ -3430,51 +3428,27 @@ public final class ReaGhidraBridge extends HeadlessScript {
         return omitted[0];
     }
 
-    private List<String> noReturnCallLimitations(
+    private List<String> functionBoundaryLimitations(
             Function function, Iterable<Instruction> instructions)
             throws Exception {
-        List<String> limitations = new ArrayList<>();
-        Instruction beforePrevious = null;
-        Instruction previous = null;
-        for (Instruction instruction : instructions) {
-            monitor.checkCancelled();
-            String limitation = terminalCallLimitation(instruction);
-            if (limitation != null) limitations.add(limitation);
-            if (beforePrevious != null && previous != null) {
-                String seh = borlandSehHandlerLimitation(
-                    function, beforePrevious, previous, instruction);
-                if (seh != null) limitations.add(seh);
-            }
-            beforePrevious = previous;
-            previous = instruction;
-        }
-        return limitations;
+        FunctionBoundaryLimitations boundary = new FunctionBoundaryLimitations(function);
+        for (Instruction instruction : instructions) boundary.observe(instruction);
+        return boundary.values();
     }
 
     private CallBoundaryObservation callBoundaryObservation(
             Function function, Iterable<Instruction> instructions, Set<Function> returnedCallees,
             boolean includeUnresolvedSites, List<Reference> observedReferences)
             throws Exception {
-        List<String> limitations = new ArrayList<>();
+        FunctionBoundaryLimitations boundary = new FunctionBoundaryLimitations(function);
         CallBoundaryReferences calls = new CallBoundaryReferences(returnedCallees, includeUnresolvedSites);
-        Instruction beforePrevious = null;
-        Instruction previous = null;
         if (observedReferences == null) visitReferences(function, "outgoing", calls::observe);
         else for (Reference reference : observedReferences) {
             monitor.checkCancelled();
             calls.observe(reference);
         }
         for (Instruction instruction : instructions) {
-            monitor.checkCancelled();
-            String limitation = terminalCallLimitation(instruction);
-            if (limitation != null) limitations.add(limitation);
-            if (beforePrevious != null && previous != null) {
-                String seh = borlandSehHandlerLimitation(
-                    function, beforePrevious, previous, instruction);
-                if (seh != null) limitations.add(seh);
-            }
-            beforePrevious = previous;
-            previous = instruction;
+            boundary.observe(instruction);
             if (!instruction.getFlowType().isCall()) continue;
             boolean destination = false;
             for (Reference reference : instruction.getReferencesFrom()) {
@@ -3488,10 +3462,38 @@ public final class ReaGhidraBridge extends HeadlessScript {
                     "Instruction decoder identifies a call but ReferenceManager supplies no call destination");
             }
         }
-        return new CallBoundaryObservation(List.copyOf(limitations),
+        return new CallBoundaryObservation(boundary.values(),
             returnedCallees == null ? null : calls.unrepresented, calls.unresolvedCalls());
     }
 
+    private final class FunctionBoundaryLimitations {
+        private final Function function;
+        private final List<String> limitations = new ArrayList<>();
+        private Instruction beforePrevious;
+        private Instruction previous;
+
+        private FunctionBoundaryLimitations(Function function) {
+            this.function = function;
+        }
+
+        private void observe(Instruction instruction) throws Exception {
+            monitor.checkCancelled();
+            String limitation = terminalCallLimitation(instruction);
+            if (limitation != null) limitations.add(limitation);
+            if (beforePrevious != null && previous != null) {
+                String seh = borlandSehHandlerLimitation(
+                    function, beforePrevious, previous, instruction);
+                if (seh != null) limitations.add(seh);
+            }
+            beforePrevious = previous;
+            previous = instruction;
+        }
+
+        private List<String> values() {
+            return List.copyOf(limitations);
+        }
+    }
+
     private String borlandSehHandlerLimitation(
             Function function, Instruction handlerPush, Instruction framePush,
             Instruction frameLink) {
@@ -3673,24 +3675,36 @@ public final class ReaGhidraBridge extends HeadlessScript {
             monitor.checkCancelled();
             Address address = addresses.next();
             if (outgoing) {
-                for (Reference reference : currentProgram.getReferenceManager()
-                        .getReferencesFrom(address)) {
-                    monitor.checkCancelled();
-                    if (!reference.isEntryPointReference()) visitor.accept(reference);
-                }
+                visitObservedReferencesFrom(address, visitor);
             }
             else {
-                ReferenceIterator iterator = currentProgram.getReferenceManager()
-                    .getReferencesTo(address);
-                while (iterator.hasNext()) {
-                    monitor.checkCancelled();
-                    Reference reference = iterator.next();
-                    if (!reference.isEntryPointReference()) visitor.accept(reference);
-                }
+                visitObservedReferencesTo(address, visitor);
             }
         }
     }
 
+    private void visitObservedReferencesFrom(Address address, ReferenceVisitor visitor)
+            throws Exception {
+        for (Reference reference : currentProgram.getReferenceManager().getReferencesFrom(address)) {
+            monitor.checkCancelled();
+            visitObservedReference(reference, visitor);
+        }
+    }
+
+    private void visitObservedReferencesTo(Address address, ReferenceVisitor visitor)
+            throws Exception {
+        ReferenceIterator references = currentProgram.getReferenceManager().getReferencesTo(address);
+        while (references.hasNext()) {
+            monitor.checkCancelled();
+            visitObservedReference(references.next(), visitor);
+        }
+    }
+
+    private static void visitObservedReference(Reference reference, ReferenceVisitor visitor)
+            throws Exception {
+        if (!reference.isEntryPointReference()) visitor.accept(reference);
+    }
+
     private static void addReference(TreeMap<String, Reference> observed, Reference reference) {
         String key = canonicalAddress(reference.getFromAddress()) + "\u0000" +
             canonicalAddress(reference.getToAddress()) + "\u0000" +

docs/browser-observation.md

@@ -281,6 +281,32 @@ REA retains local observation data and removes only URL userinfo credentials, ex
 - Accessibility structure and roles are retained by default, while names and descriptions require `include_accessibility_text: true`.
 - Network observations retain method, status, MIME type, size, type, initiator stack location, and complete URLs with userinfo credentials removed. Redirects reported by CDP are attached as ordered `redirects` on the one request ID, with the prior request URL, CDP response URL, method/type, response status/MIME/encoded length, request timestamp, and the next `requestWillBeSent` event timestamp. CDP delivers `redirectResponse` with that next request event, so the latter is an event boundary rather than an exact response timestamp. Redirect response headers and bodies are discarded; hops outside the selected origins are discarded with the request chain. Headers are discarded after an allowlisted projection of length/encoding, structured CSP/Link/policy fields, and untrusted agent hints. Cookies and authorization headers are never retained.
 - Request/response bodies are not requested or parsed by default. When selected, only allowed-origin JSON media types are read, converted immediately to complete value-free typed paths, types and observation counts, and discarded. Each path segment is either `{ "kind": "property", "name": "raw property name" }` or `{ "kind": "array-element" }`; array-element types include primitives and nested arrays. Literal `*` properties remain distinct from array elements. `Network.getResponseBody` is never sent unless JSON body shape capture is selected. Comparison uses body shapes only when both captures selected them, and compares request/response shapes only where both observations provide those facts. Unavailable or truncated selected coverage makes equivalence unknown; observed differences still report changed.
+  For normalized passive inputs with complete request inventories on both sides,
+  request additions and removals remain observable even when selected body shapes,
+  response metadata, or WebSocket coverage are unavailable. HTTP request and
+  WebSocket inventory coverage are evaluated separately; complete HTTP coverage
+  cannot prove a missing WebSocket connection. Limited request inventory still
+  prevents request-absence claims. This comparison rule does not make live CDP's
+  attach-limited inventory complete or establish body-shape equality.
+  An unavailable HTTP resource type also leaves request identity incomplete:
+  observations with the same method and URL remain possible matches across types,
+  and identical unknown types cannot establish complete equality. Ambiguous type
+  groups cannot prove membership or content changes either: a missing type may
+  redistribute repeated requests between buckets without changing the requests.
+  Their comparison remains unknown, while independently identified groups retain
+  proven changes. Different methods/URLs and complete empty peer inventories still
+  support absence claims.
+  Relevant dropped-event counts also limit inventory completeness even when no
+  section is listed as truncated: HTTP uses `network_requests`, while WebSocket
+  uses `websocket_connections` and `websocket_frames`. Drops in other event kinds
+  and the aggregate total do not by themselves limit HTTP inventory. Proven
+  modifications remain reportable despite incomplete inventory.
+  Repeated identical HTTP content samples establish multiplicity changes only
+  with complete request inventories; limited inventories retain known content
+  comparisons without interpreting duplicate sample counts as changes.
+  WebSocket stream inventories and connection counts likewise require complete
+  WebSocket coverage. Redirect resource types use only shared known metadata:
+  an omitted type leaves equality unknown rather than proving a modification.
 - Console observations with a stack source retain call type, argument types, timestamp, and source location. When console text capture is selected, already-delivered primitive values are retained verbatim; objects, getters, and remote properties are never expanded.
 - WebSocket observations retain direction, opcode, and payload byte length. When shape capture is selected, text frames are classified as text or complete value-free JSON shape; binary bytes, hashes, prefixes, and raw frames are never retained.
 - Storage observations always redact values. Key names, IndexedDB names, and cache names require `include_storage_keys`; stable content fingerprints require the additional `include_storage_fingerprints` selection. Cookie fingerprints cover cookies applicable to the current main-frame URL, including its path, rather than every cookie path on that origin. Other storage remains scoped to the origin. Cache bodies above 64 KiB and partial IndexedDB remote objects make fingerprint coverage incomplete, so identical observations remain `unknown`.
@@ -317,8 +343,10 @@ an invalid supplied size remains unknown with incomplete resource coverage while
 the resource's other facts are retained. Zero and nonnegative fractional sizes
 are preserved.
 
-CDP discovery, connection, and commands wait for the peer or caller cancellation;
-a stalled peer can leave an operation waiting. WebSocket messages remain subject
+CDP discovery has a 10-second deadline. Each WebSocket handshake and command
+response has its own 20-second deadline, separate from the caller-selected
+observation window. A stalled handshake or command reports `provider_timeout`;
+caller cancellation still reports cancellation. WebSocket messages remain subject
 to the [64 MiB limit](#request-boundary). REA does not impose a message-count or
 pending-command quota. On cancellation, it closes only its own WebSocket,
 waiting up to one second before terminating it. Malformed unsolicited events

docs/browser-scenario-contract.md

@@ -23,9 +23,9 @@ that rejection does not roll back browser effects. REA removes its owned
 future-document scripts, restores live frame methods, then disconnects.
 Add secret declarations, storage seeds, and additional artifact or event
 capture only when needed. Action, secret, and storage counts are not capped.
-Duration, action, and navigation
-timeouts remain fixed provider-owned liveness limits; the request does not
-accept a caller-controlled `limits` object.
+Duration, action, and navigation timeouts remain fixed provider-owned liveness
+limits. `capture.limits` controls retained event and network-content budgets;
+it does not change those liveness deadlines.
 
 ## Minimal launch example
 
@@ -179,12 +179,44 @@ declared action. Each step reports action status, elapsed time, sanitized URLs,
 event bounds, and independently typed capture states for screenshot, DOM,
 accessibility, URL, history, and storage. Console, page-error, network,
 WebSocket, frame, worker, popup, and cancelled-download events are attributed to
-steps and all observed events and requested artifacts are returned inline.
-There are no event, frame, worker, popup, WebSocket, DOM-node,
-accessibility-node, storage-entry, screenshot-count, or cumulative
-metadata-byte caps. Requested artifacts and observed events are returned
-inline without application-defined size ceilings. Other missing sections
-remain explicit and make the capture ineligible for equality claims.
+steps. Retained events and requested artifacts are returned inline; omitted
+events are counted by kind and identify affected network transactions.
+Event and selected network-content retention have caller-selected limits under
+`capture.limits`. Omitted values use the defaults below; every supplied value
+must be a positive safe integer, and callers can raise or lower a default.
+Byte limits are conservative allocation-cost estimates for retained event
+values, including UTF-16 storage and JSON expansion, rather than serialized
+bytes or a promise about exact process memory. `events.omissions.by_kind`
+reports counts and reasons for omitted event kinds. Its
+`network_transactions` rows identify affected transaction IDs and omitted
+event kinds; `network_transaction_range` identifies the contiguous IDs beyond
+the transaction limit. Omission metadata is bounded by the selected
+transaction limit. Other missing sections remain explicit and make the capture
+ineligible for equality claims.
+
+The default limits are 10,000 events, 4 MiB estimated allocation for one
+event, 16 MiB estimated allocation across retained events, 1,024 network
+transactions, 64 KiB of body bytes per transaction phase, 1 MiB estimated
+allocation for body redaction/base64, 1 MiB for retained header values, and
+1 MiB for network metadata. The same limits are included in the capture
+Evidence. Playwright materializes response bodies before REA can inspect their
+byte length; `max_network_body_bytes` bounds subsequent retained copies and
+does not bound the browser API read itself. A body that exceeds its selected
+limit has an explicit `body-size-limit` or `body-allocation-limit` unavailable
+state with the observed and allowed values.
+
+Additional implementation bounds are 16 simultaneous network-content reads,
+a 5-second deadline for each provider API read, and 2,048 live event-listener
+owners. These are not caller-selectable. The concurrency bound queues admitted
+content jobs and does not itself omit observations; a provider read timeout
+produces a typed unavailable state, while exceeding listener capacity produces
+event-kind omission counts. A preflight avoids body redaction and base64 copies
+when the currently available event budget already rejects the body.
+Concurrent reads may each pass that check before their events are retained;
+their retained-copy work is bounded by 16 times the selected per-body
+allocation limit. The Playwright response Buffer itself is materialized before
+REA can apply that limit. Requested artifacts and retained events are returned
+inline within the declared limits.
 Attach-mode captures also declare the unavoidable pre-attach event gap.
 
 Selected event families are observed on the root page and discovered descendant
@@ -288,16 +320,21 @@ retained representation after redaction.
 
 Only completed responses are read. At the end of the final step, observation
 stops; requests still in flight emit `request-unfinished`, and their observed
-responses have explicitly unavailable bodies. This cutoff does not wait for
-long-lived streaming connections. Already-started content reads settle before
-browser cleanup, with a provider-owned five-second timeout per read and
-cancellation support. Failed or unexposed reads retain per-content states and make selected
-event coverage incomplete. Body events can arrive during later steps or final
-drain; event bounds describe receipt during the step, while the source event
-reference establishes the transaction phase. Comparison attributes content, like
-every other event of its transaction, to the step that sent the request (the
-receipt step when that request was not observed) and excludes the sequence
-reference from normalization.
+responses have explicitly unavailable bodies. Already-admitted content reads
+drain before browser cleanup. At most 16 reads run at once; queued work is
+bounded by the selected network-transaction limit and by successful admission
+of its request or response event. Each provider API read has its own five-second
+timeout once it starts, so queue wait adds to total drain and cleanup time; this
+is not a five-second total capture cutoff. Cancellation interrupts active reads
+and queued jobs produce cancelled states without calling their provider APIs.
+This drain does not wait for long-lived unfinished streaming connections and
+does not refetch unfinished response bodies. Failed or unexposed reads retain
+per-content states and make selected event coverage incomplete. Body events can
+arrive during later steps or final drain; event bounds describe receipt during
+the step, while the source event reference establishes the transaction phase.
+Comparison attributes content, like every other event of its transaction, to
+the step that sent the request (the receipt step when that request was not
+observed) and excludes the sequence reference from normalization.
 Playwright scenario capture does not supply initiator stacks; passive CDP
 observation retains its own initiator evidence.
 

docs/cli.md

@@ -314,16 +314,56 @@ diagnostics appear inside the report when `--full-output` is selected. Log
 truncation alone does not change the diagnostic health or exit status.
 
 The default terminal format is TOON. Use `--json` when saving results for a
-JSON consumer. `--json` is indented on a terminal and compact when piped or
+JSON consumer. Valued output flags (`--format`, `--filter-output`,
+`--token-limit`, `--token-offset`) accept `--flag value` or `--flag=value`; a
+missing value fails with `MISSING_OUTPUT_VALUE` before the command runs. `--json` is indented on a terminal and compact when piped or
 redirected, which keeps large results smaller for agents and scripts; pipe
 through `jq .` for indented files. Output selection and formatting do not change
 operation status.
 
-| Status    | Meaning                                                                                                                                              |
-| --------- | ---------------------------------------------------------------------------------------------------------------------------------------------------- |
-| `0`       | The operation completed. Its result may include partial evidence, warnings or unresolved questions.                                                  |
-| `1`       | The operation could not complete. Structured output identifies invalid input, permissions, cancellation, timeouts or another failure when available. |
-| `128 + N` | Signal `N` ended the process, where the shell or runtime preserves the conventional signal-derived status.                                           |
+For `cutter-command`, `--json-output` requests command-specific JSON from
+Cutter when available. This is separate from `--format json`, which selects the
+CLI result representation; the global `--json` flag selects that same CLI
+representation.
+
+Cutter still accepts deprecated `--json=true` and `--json=false` compatibility
+values. When both forms are present, `--json-output` takes precedence in either
+argument order. Bare `--json` retains its global meaning of selecting JSON CLI
+output formatting.
+
+| Status    | Meaning                                                                                                                                                                                                                                                                                                                                                                                |
+| --------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
+| `0`       | The operation completed. Its result may include partial evidence, warnings or unresolved questions.                                                                                                                                                                                                                                                                                    |
+| `1`       | The operation could not complete. Structured output identifies invalid input, permissions, cancellation, timeouts or another failure when available.                                                                                                                                                                                                                                   |
+| `128 + N` | Signal-derived status. An unhandled signal may have ended the process, with the shell or runtime preserving the conventional status; a command may instead have handled the signal, waited for its operation and cleanup to settle, and assigned the same status (`130` for SIGINT, `143` for SIGTERM). In the handled case, it may also have printed a structured `cancelled` result. |
+
+A `cancelled` result does not by itself determine the exit status; it depends on
+the command's signal handling. Commands routed through the shared
+`withCommandCancellation` handler in the CLI set status `130` for SIGINT or
+`143` for SIGTERM after their operation settles. Current source registrations
+use that handler for process capture and comparison (`capture-process`,
+`compare-process-captures`), JavaScript application analysis (`analyze` when it
+selects the JavaScript lane, and `analyze-javascript-application`), selected
+browser, web and Electron commands (including `analyze-web-bundle`,
+`observe-web-execution`, `capture-browser-scenario` and
+`capture-electron-scenario`), and the platform and provider command families
+for Android/ADB, Apktool/JEB/Flutter, firmware, PCAP and signal analysis,
+WASM/EVM, Go/Rust and binary diagnostics, Frida instrumentation, the Cutter
+command, objdump, Rizin and GDB. In these paths, a returned structured
+`cancelled` result can be printed before the wrapper assigns the
+signal-derived status. This describes the current source routing; it is not a
+claim that every command or provider runtime has been independently exercised.
+
+When the reader of piped output closes early, as `rea ... | head` does, the
+CLI stops writing, prints no error and exits `141` (`128 + 13`, the status of a
+SIGPIPE-terminated writer). The operation itself may have completed; only its
+output was declined.
+
+`inspect-artifact` also returns a structured cancelled result, but its direct
+analysis cancellation path does not set a signal-derived exit status, so it
+exits `1`. Setup has its own contract: cancellation exits `0`. Treat the
+structured result and process status as separate signals when scripting
+cancellation.
 
 `setup --dry-run` returns `planned` and exits `0`; a cancelled setup also exits
 `0`. Setup returns `1` for `needs_confirmation` or `needs_human`.

docs/client-configuration.md

@@ -140,8 +140,35 @@ profiles](mcp-contracts.md#output-schemas).
 Setup registers `rea` in `~/.kimi-code/mcp.json`, or
 `$KIMI_CODE_HOME/mcp.json` when set, while preserving unrelated settings and
 creating a `.rea.backup` before updating an existing file. Kimi Code infers
-stdio from the `command` entry, so setup does not add a `type` field. The
-registration includes `startupTimeoutMs = 30000` and `toolTimeoutMs = 120000`.
+stdio from the `command` entry, so setup does not add a `type` field. New
+registrations include `startupTimeoutMs = 30000` and a `toolTimeoutMs` derived
+from REA's known provider allowance plus 120 seconds of operation
+headroom. The default is 450000 ms, covering the default 330-second Ghidra
+startup; an explicit Ghidra startup timeout of 2700000 ms produces a 2820000 ms
+tool timeout. The startup allowance follows an explicitly selected Ghidra or
+Hopper provider, and otherwise uses the larger known Ghidra/Hopper deadline.
+An explicit `REA_JEB_MCP_TIMEOUT_MS` request budget also contributes to that
+maximum. For `ida` or `auto` selection, a valid IDA registration's parsed
+`timeoutMs` also contributes to that maximum; its default is 300000 ms.
+Setup forwards these explicitly selected REA deadlines and the input schema
+profile into the registration; unrelated process values are omitted.
+This headroom is a practical default, not a guarantee for every analysis
+duration. Setup preserves positive timeout values already in the `rea` entry.
+You can edit `startupTimeoutMs` and `toolTimeoutMs` there; doctor keeps positive
+caller-selected values aligned and reports a non-blocking advisory when a tool
+timeout is shorter than the known provider allowance.
+Setup selects the compact input-schema advertisement for new or refreshed Kimi
+registrations unless their stored configuration or setup environment explicitly
+selects `REA_MCP_INPUT_SCHEMA_PROFILE=full` or `compact`. This keeps REA's
+canonical tool validation unchanged while avoiding a Moonshot Kimi Code 2.2.0
+failure observed with the full catalog: its request was rejected because a
+single tool schema exceeded the provider's size limit. The limit is
+provider-specific and not an MCP protocol limit. `rea doctor --client kimi`
+reports an older registration with no explicit profile as stale. To restore the
+compact profile, run `rea setup --client kimi` or add
+`REA_MCP_INPUT_SCHEMA_PROFILE=compact` to the registration's `env`, then restart
+Kimi Code. An explicit `full` profile remains respected and can still exceed
+Moonshot's limit.
 Kimi Code also discovers personal skills in `~/.agents/skills` and project-local
 `.agents/skills`, so no separate skill root is required; setup still installs
 and verifies the matching skill files through the normal shared-skill flow.
@@ -207,8 +234,20 @@ standalone CLI check.
 
 ## Codex and Devin
 
-Codex treats an unset or empty `CODEX_HOME` as `~/.codex`. Devin honors
-`XDG_CONFIG_HOME` for its `devin/mcp_config.json` on non-Windows systems.
+Codex treats an unset or empty `CODEX_HOME` as `~/.codex`. Its MCP server
+registration supports `startup_timeout_sec` for connection initialization and
+`tool_timeout_sec` for each tool call. New REA registrations use a 30-second
+startup timeout and a tool timeout derived from the known provider
+allowance plus 120 seconds of operation headroom. This yields 450 seconds for
+the default Ghidra allowance; an explicit Ghidra startup timeout of 2700
+seconds yields 2820 seconds. Setup preserves positive caller-selected timeout
+values in the REA entry, and doctor reports a non-blocking advisory when a tool
+timeout is shorter than the known provider allowance. An explicitly selected
+JEB request budget also contributes to the allowance. A configured IDA
+registration's parsed `timeoutMs` contributes for `ida` or `auto` selection.
+These values are client-side deadlines, so increase them in `~/.codex/config.toml` if a
+workload needs more time. Devin honors `XDG_CONFIG_HOME` for its
+`devin/mcp_config.json` on non-Windows systems.
 
 ## OMP
 

docs/client-verification.md

@@ -6,6 +6,28 @@ and provider lanes. Each record below states its host, client version, model
 fixture, and remaining coverage; a native-client run with a loopback model does
 not establish live model-service acceptance.
 
+## Published Pi code-mode renderer limit
+
+A separate probe of the published `@earendil-works/pi-codemode` 1.1.0 and 1.2.0
+declaration renderers rendered the full `compare_web_captures` input as
+`args: unknown`: its generated TypeScript type exceeds Pi's 16,000-character
+default. REA keeps the full advertisement's fields by inlining references when
+the shared form would exceed Pi's 32-reference expansion limit; retaining that
+shared form would leave later fields unknown instead. The existing
+`REA_MCP_INPUT_SCHEMA_PROFILE=compact` profile lets callers stay below
+its character limit by advertising reduced nested validation structure, while
+REA continues to validate calls against the complete canonical input schema.
+Compact applies to the server's advertised inputs and is not an automatic
+full-profile fallback; see the [compact input schema profile](mcp-contracts.md#compact-input-schema-profile)
+for details of its reduced forms.
+
+Custom renderer hosts can raise `inputMaxChars` through `renderToolSignature` or
+`renderToolSample`; `renderDeclarations` does not expose that override. The
+probe exercised the published renderer code, not a native Pi host or model, and
+does not establish broader Pi compatibility. See the pinned upstream
+[`declarations.ts`](https://github.com/earendil-works/pi/blob/b45e0ebcc5ae2fc42a6f2f23104d10e656e6bedb/packages/codemode/src/declarations.ts)
+and [renderer usage notes](https://github.com/earendil-works/pi/blob/b45e0ebcc5ae2fc42a6f2f23104d10e656e6bedb/packages/codemode/README.md#declarations-for-the-model).
+
 ## End-to-end, integration and golden evidence
 
 The optional `verify:qwen-client`, `verify:pi-client`, and `verify:hermes-client`

docs/electron-observation.md

@@ -116,6 +116,12 @@ If cleanup cannot be verified, the error retains collected observations in
 `details.partial_observation` and marks their application cleanup `unverified`.
 Cleanup uncertainty does not authorize signaling unrelated processes.
 
+### Target fuse prerequisites
+
+Active scenario capture requires that the target Electron executable honors Node CLI inspector flags (`--inspect` or `--remote-debugging-port`) so that Playwright can establish its DevTools protocol session.
+
+When an Electron executable explicitly disables the `EnableNodeCliInspectArguments` fuse (or removes the capability), `capture-electron-scenario` fails immediately before process launch with an `AnalysisUnsupportedTargetError` ("unsupported_target") rather than waiting for the startup deadline. For targets with disabled inspector fuses, use passive CDP inspection via `inspect_electron_page` or run against an unrestricted development binary with the extracted `app.asar`.
+
 The real verifier exercises compiled CLI and stdio MCP discovery, page/script
 inspection, static/runtime reconciliation, active observation, and CLI SIGTERM
 cleanup before checking the IPC, renderer crash/reload, utility/child-process,

docs/ghidra-provider.md

@@ -153,8 +153,8 @@ overview merely because its source artifact is unchanged.
 
 Function extents and instruction completeness refer to Ghidra's analyzed body.
 An inferred no-return call can terminate that body before subsequent bytes.
-Pseudocode, assembly, procedure info, instruction windows, callee and caller
-reads, and dossiers report observed terminal call sites in Evidence limitations,
+Pseudocode, assembly, procedure info, instruction windows, incoming and outgoing
+references, callee and caller reads, and dossiers report observed terminal call sites in Evidence limitations,
 including resolved callee names, addresses, and FunctionManager flags when
 available. On x86-32, the same views also report the exact Borland/Delphi-style
 FS exception-registration pattern when its handler address falls outside the

docs/ida-provider.md

@@ -131,9 +131,9 @@ Use a package version that actually includes this adapter; repository main can b
 {"name":"close_binary","arguments":{}}
 ```
 
-Opening an MCP target starts its selected provider session; headless opening waits for auto-analysis. Set the agent's MCP request deadline long enough for the target. The registration's optional `timeoutMs` defaults to 300000 for upstream requests. A freshly opened headless worker can briefly report `status: "busy"` from `server_health`; REA re-reads the health every 100 ms until the worker answers, the caller cancels, or `timeoutMs` has elapsed since the first read. Only expiry while the worker is still busy produces a retryable `provider_timeout`; no further read is started once the deadline has passed. That error names the busy tool of the last observation, or `unknown` when the worker reported none or a value that is not tool-name-shaped. Cancellation during the wait produces `cancelled`. Any other unexpected health payload, including a busy payload with an invalid shape, remains `unreadable_output`. Cancellation drains the current upstream request before cleanup; it does not force-stop GUI analysis. An open request that times out without a completed response retains the private workspace because worker inventory alone cannot prove the request has stopped. `binary_session` reports actual tool availability. Repeated doctor/setup calls are unnecessary when the registration already works.
+Opening an MCP target starts its selected provider session; headless opening waits for auto-analysis. Set the agent's MCP request deadline long enough for the target. The registration's optional `timeoutMs` defaults to 300000 for upstream requests. An upstream SDK request that reaches this limit returns `provider_timeout` with its operation and actual request budget. A freshly opened headless worker can briefly report `status: "busy"` from `server_health`; REA re-reads health every 100 ms until the worker answers, the caller cancels, or `timeoutMs` has elapsed since the first read. Each health request is capped by the time remaining in that overall wait, so a stalled read cannot extend the readiness deadline by another full request timeout. Expiry while the worker is still busy produces a retryable `provider_timeout`; no further read is started after the deadline. That error names the busy tool from the last observation, or `unknown` when the worker reported none or a value that is not tool-name-shaped. If a health request itself consumes the remaining wait budget, the same last-busy timeout is returned with the stalled request's captured output. Cancellation during the wait produces `cancelled`. Any other unexpected health payload, including a busy payload with an invalid shape, remains `unreadable_output`. Cancellation drains the current upstream request before cleanup; it does not force-stop GUI analysis. An open request that times out without a completed response retains the private workspace because worker inventory alone cannot prove the request has stopped. `binary_session` reports actual tool availability. Repeated doctor/setup calls are unnecessary when the registration already works.
 
-For guided registration, set `REA_IDA_MCP_CONFIG` before `rea setup --client <client>`. Setup includes that file reference in its reviewed plan and preserves it after approval; it does not copy upstream credentials into the agent configuration. `rea doctor --provider ida --json` validates the registration without starting IDA and explicitly leaves runtime compatibility and target binding for analysis.
+For guided registration, set `REA_IDA_MCP_CONFIG` before `rea setup --client <client>`. Setup includes that file reference in its reviewed plan and preserves it after approval; it does not copy upstream credentials into the agent configuration. For `ida` or `auto` selection, Codex/Kimi setup and doctor use the effective client registration's parsed `timeoutMs` (default 300000 ms) when recommending the agent's per-tool deadline. Existing positive client deadlines remain caller choices. `rea doctor --provider ida --json` validates the registration without starting IDA and explicitly leaves runtime compatibility and target binding for analysis.
 
 ## Coverage and evidence
 

docs/installation.md

@@ -147,7 +147,8 @@ identity remains available at `identity.catalog`.
 
 Setup pins package-runner MCP registrations to the exact installed REA version,
 installs the matching skill and on-demand references in the same plan, and adds
-`startup_timeout_sec = 30` for Codex and Grok Build.
+`startup_timeout_sec = 30` for Codex and Grok Build. Codex also receives a
+`tool_timeout_sec` derived from known provider startup plus operation headroom.
 
 ## Limit REA to one project or agent profile
 
@@ -397,8 +398,9 @@ running server or its registration.
 The npm release checked on **2026-10-11** was **6.4.0**, published from
 [`a56f7bb6`](https://github.com/morluto/rea/commit/a56f7bb668ab23fec8d33f7af353ac7ed2fe1895).
 The actual registry package accepted dry-run setup plans for the 17 client IDs
-in that release, including Qwen Code, Pi and Hermes. DeepSeek Harness (`dsh`)
-is included from 6.5.0 and was absent from the checked npm 6.4.0 package.
+in that release, including Qwen Code, Pi and Hermes. Seven client IDs are
+included from 6.5.0 and were absent from the checked npm 6.4.0 package: `droid`,
+`dsh`, `kimi`, `qoder`, `trae`, `workbuddy` and `zcode`.
 The 6.4.0 registry check does not establish 6.5.0 publication or live DeepSeek
 API integration.
 OpenCode's plan selected
@@ -481,8 +483,8 @@ hosted connector; see [its configuration](client-configuration.md#other-client-s
 | ZCode              | `zcode`          |
 | DeepSeek Harness   | `dsh`            |
 
-`dsh`, `kimi`, and `workbuddy` require a release containing them. The registry
-check above does not cover them.
+`droid`, `dsh`, `kimi`, `qoder`, `trae`, `workbuddy`, and `zcode` require
+`rea-agents@6.5.0` or later. The 6.4.0 registry check above does not cover them.
 
 See [client configuration and troubleshooting](client-configuration.md) for
 configuration paths, profiles, trust policies, and schema-limit workarounds.
@@ -540,6 +542,37 @@ search permissions (`0700`) on a directory you own before retrying.
 
 On supported Linux distributions, approved setup verifies Hopper's official
 `.deb`, `.rpm`, or Arch package before invoking the native package manager.
+The official release feed selects the current public demo without a version
+pin. REA verifies the package's detached PGP signature against Hopper's stable
+primary fingerprint `7C2D73BAFAF5B9A710A69DA1A26DF815099B1BCD`, then checks
+that the signed package and installed launcher agree with `main_exe_hash`.
+Rotating signing subkeys remain bound to that primary identity.
+The feed row itself is not PGP-signed; doctor labels its display version as
+`reported_release_version`, separately from the authenticated launcher identity.
+
+The approval plan includes the caller-selected cache location. Approved setup
+retains the signed archive, public key and release metadata under
+`${XDG_CACHE_HOME:-$HOME/.cache}/rea/hopper-releases/<launcher-sha256>`.
+An empty `XDG_CACHE_HOME` uses `$HOME/.cache`. Approved setup can repair a
+damaged caller-owned entry after authenticating a fresh package; the previous
+entry is preserved beside it as `<launcher-sha256>.backup-<id>`. Ordinary doctor
+and launches reject damaged evidence and never repair it implicitly.
+Approved publications of the same canonical cache entry share a process-backed
+lease. Concurrent setup waits without moving another writer's evidence; a
+failed replacement restores the previous directory when possible and reports
+any retained backup path. Unavailable reads or native inspection tools do not
+authorize replacing existing evidence.
+Doctor and launches authenticate this evidence for offline use; an editable
+digest receipt alone cannot authorize a launcher. Verification reuses at most
+eight immutable package proofs per process, keyed by the complete package,
+key, signature, package format and expected launcher digest. Cached file reads
+still check identity and content. Without retained evidence, inspecting an
+existing install requires the matching current official package and network
+access. Successful live inspection can be reused within that process after
+rehashing the launcher; it does not write a persistent cache or claim that a
+previously observed release is still the latest. Older uncached builds require
+an approved update while the feed exposes only its current release.
+
 REA runs the supported demo build on a private Xvfb display and selects Hopper's
 offered demo mode for each analysis session; it does not require the user's
 desktop display. Unattended package-manager access requires
@@ -634,6 +667,20 @@ If startup reports `Cannot launch Hopper! Connection is invalid.` or waits for
 bridge readiness, check Hopper for a pending demo, license, or loader dialog.
 Complete the dialog and retry. If REA reports `cleanup_incomplete`, follow the
 retained-resource guidance below before reopening the target.
+
+On macOS, Hopper can show an `Executing Python Script...` overlay while the REA
+bridge serves an active MCP connection. Long-lived MCP clients can keep that
+connection open across agent turns, so the overlay may remain visible after the
+analysis calls finish; by itself, it does not establish a stalled startup. REA
+does not automatically idle-close a still-connected target. When the
+investigation is finished, export an Evidence bundle first if you need to keep
+the session ledger, then call `close_binary`; set `snapshot_path` to save a
+provider-neutral analysis snapshot as part of closing. A successful close
+shuts down the REA bridge and clears its overlay; REA does not control the
+macOS Hopper application lifetime.
+The Hopper GUI Cancel button's behavior for this bridge has not been verified;
+`close_binary` is the supported cleanup path.
+
 Hopper serializes analysis requests. Cancelling a wait can leave provider work
 running, which the session reports. Successful decompilation is cached until
 a relevant rename or comment changes it.

docs/javascript-application-workflows.md

@@ -60,6 +60,8 @@ provider implicitly.
 `trace_javascript_semantics` queries the authenticated companion graph without
 rereading mutable files. A seed may identify a semantic or application node,
 literal, function fingerprint, property, endpoint, event, or boundary field.
+For a recovered parse, coercion, or validation boundary, its field value is the
+static operation's method name (for example, `Number`).
 The query declares backward provenance, forward influence, callers, or
 ownership and can restrict admitted relation kinds. REA returns every matching
 seed and the complete reachable subgraph; callers do not choose node, relation,
@@ -83,6 +85,11 @@ callee bodies and built-ins such as `Object.assign`, `Reflect.set`, and
 values. This uncertainty does not prove that a call actually mutated a value.
 Interprocedural aliasing beyond the supported call-site references remains
 outside this mutation pass.
+For process configuration, only known environment value slots and canonical
+argument indexes are treated as primitive when passed to calls. Unknown or
+inherited selections can expose a mutable prototype, so later reads of that
+configuration container retain uncertainty; earlier top-level reads and unrelated
+configuration containers remain independent.
 
 Function fingerprints commit normalized syntax, control-flow shape, relation
 shape, literal sets, arity, and detected effects without using local names or

docs/javascript-artifact-reconstruction.md

@@ -206,6 +206,9 @@ code unit, matching the original bytes identified by the artifact digest.
 
 Plain `.ts` artifact sources use TypeScript syntax without JSX, including
 angle-bracket type assertions. `.tsx` and `.jsx` retain JSX parsing.
+JavaScript sources use JavaScript grammar with JSX, never TypeScript syntax,
+so valid JavaScript that TypeScript reads differently keeps its JavaScript
+meaning.
 
 Relative TypeScript imports can name their emitted `.js`, `.mjs`, or `.cjs`
 counterparts while resolving to inventoried source files. This substitution
@@ -220,7 +223,10 @@ Selected bounded text is then parsed as inert data to recover:
 - Webpack and Rspack chunk registrations and module factories represented as
   AST literals;
 - static imports, dynamic imports, CommonJS `require` calls, workers, and
-  service workers;
+  service workers (only `new Worker(new URL(path, import.meta.url))` resolves
+  relative to the calling module; string worker and service-worker paths use
+  the document base URL or Node's working directory, which static analysis
+  does not establish, so they keep their expression and stay unresolved);
 - route, network endpoint, storage, and vendor-marker observations;
 - local source-map declarations and original
   source names and content digests;
@@ -626,6 +632,10 @@ href are read without surrounding whitespace or embedded tabs and newlines, and
 their percent-encoded path bytes are decoded; encoded dot and separator bytes
 are still rejected. CommonJS module lookups retain extension and directory
 resolution.
+ESM, HTML, and URL references apply URL dot segments to empty path segments:
+`./assets//../app.js` selects `assets/app.js`, while a remaining empty segment
+such as `./assets//app.js` stays unavailable. CommonJS keeps filesystem path
+semantics and selects `app.js` for the same text.
 Unresolved HTML references retain their declaration, source range, and resolution
 reason in the renderer observations.
 HTML script source ranges follow the HTML parser across LF, CRLF, and bare CR

docs/javascript-runtime-observation.md

@@ -95,6 +95,11 @@ pauses or resumes execution, calls `Runtime.runIfWaitingForDebugger`, launches
 the target, instruments JavaScript, or invokes Electron APIs. Closing a call
 closes only REA's WebSocket.
 
+Each WebSocket handshake and command response has its own 20-second deadline,
+separate from the caller-selected observation window. A stalled handshake or
+command reports `provider_timeout` and retains observations already received;
+caller cancellation still reports cancellation.
+
 The direct observations are:
 
 - `Debugger.scriptParsed`, establishing that a bounded script location was

docs/jeb-analysis.md

@@ -16,6 +16,21 @@ URL after validating its syntax. That validation does not establish that the
 JEB service is reachable or that a project is open; use the existing
 `inspect-jeb-client` command for a read-only service query.
 
+MCP `binary_session` probes the live JEB service and its advertised tool catalog.
+Availability is reported per operation: a reachable headless client can support
+inspection and decompilation while refusing `open_jeb_project`. Catalog discovery
+uses one 60-second deadline across connection and pagination; a responsive server
+cannot keep discovery running indefinitely by returning more pages.
+
+Each upstream tool call has an absolute 60-second request deadline by default.
+Set `REA_JEB_MCP_TIMEOUT_MS` to a positive decimal number of milliseconds to
+select a different per-request budget, for example `REA_JEB_MCP_TIMEOUT_MS=300000`
+for a five-minute decompilation. This budget starts after discovery and applies
+to each individual `tools/call`; it does not extend the separate catalog deadline
+or establish one end-to-end deadline across a whole CLI or MCP operation. A
+request that reaches its budget returns a typed timeout, while caller
+cancellation remains a cancellation.
+
 Verified against JEB 5.48.0 (Linux x64, licensed build) serving `jeb-mcp-server` 1.3.0 over streamable HTTP.
 
 ## Operations

docs/mcp-contracts.md

@@ -38,13 +38,10 @@ socket and device entries are not opened. Case-distinct archive members remain
 separate observations. Their case-insensitive materialization diagnostic does
 not reduce traversal coverage when both spellings were read.
 
-Current output also retains `coverage_status` (`complete` or `partial`) for existing clients, derived from the scoped traversal status. The owning parser normalizes retained `coverage_status` records to scoped coverage and rejects conflicting declarations.
-
-The coverage field is optional when reading older retained Evidence. The owning
-inventory parser recognizes historical REA scope notices for compatibility;
-unfamiliar legacy limitations remain conservative. Typed records do not derive
-coverage from limitation text. Pages from one inventory must carry the same
-coverage declaration.
+Current inventory Evidence requires scoped `coverage` with a `complete` or
+`unknown` status. Consumers use this producer-owned fact and reject records
+without it; limitations are explanatory and do not supply or infer coverage.
+Pages from one inventory must carry the same coverage declaration.
 
 Function dossiers can declare per-facet coverage in the optional
 `normalized_result.facet_availability` map. Each supplied facet is either
@@ -520,6 +517,14 @@ These deadlines have different owners:
 | Darwin swiftc helper | REA allows 120,000 ms by default for ownership-helper compilation; `REA_SWIFTC_COMPILE_TIMEOUT_MS` accepts decimal integers from 1–2,147,483,647 ms. Invalid values fail configuration validation.                                  |
 | Individual tool call | The client bounds its wait, including cold engine startup. The pinned client SDK 2.3.1 defaults to 60,000 ms and can cancel earlier than REA's startup deadline.                                                                    |
 
+For [Codex and Kimi registrations](client-configuration.md), setup supplies
+missing tool deadlines from the known provider allowance plus 120 seconds of
+operation headroom: 450 seconds with the default Ghidra allowance, or 2820
+seconds with `REA_GHIDRA_STARTUP_TIMEOUT_MS=2700000`. Positive caller-selected
+deadlines remain unchanged. Doctor reports an informational recommendation
+when a configured deadline is shorter than the known provider allowance. An
+explicit JEB upstream request budget also contributes to that allowance.
+
 For an already connected client using the pinned SDK, request options are the
 **second** argument of `callTool`:
 
@@ -549,10 +554,20 @@ Waiting notifications stop when startup settles or the caller cancels. They do
 not measure engine work completed.
 
 The pinned SDK can reset the request's inactivity timeout on progress with
-`resetTimeoutOnProgress: true`. Also set `maxTotalTimeout` to bound the complete
-request; the provider's startup deadline still applies. These options belong in
-the same second argument of `callTool`. Without that explicit client policy, the
-default 60-second request deadline still expires despite waiting notifications.
+`resetTimeoutOnProgress: true`. These options belong in the same second
+argument of `callTool`. Without that explicit client policy, the default
+60-second request deadline still expires despite waiting notifications.
+
+SDK 2.3.1 checks `maxTotalTimeout` when progress resets the timer; it does not
+schedule a separate absolute timer for that setting. A silent interval can
+therefore run beyond the declared total. For an absolute caller deadline,
+also pass `signal: AbortSignal.timeout(totalBudgetMs)`. If the caller already
+has a cancellation signal, preserve both with
+`AbortSignal.any([callerSignal, AbortSignal.timeout(totalBudgetMs)])`.
+`maxTotalTimeout` can still bound progress resets, and REA's provider startup
+deadline still applies. Choose the total budget for the expected workload;
+neither server configuration nor progress can override a remote caller's
+deadline.
 
 The 240-second setting is a measured example, not a universal timeout: a
 controlled Linux x64/WSL2 fixture took about 64 seconds for its first overview
@@ -664,10 +679,14 @@ advertised. Every advertised example is a distinct valid canonical input.
 
 Root object unions are presented as
 one object without root `anyOf`, `oneOf`, or `allOf` for model API compatibility.
-The projection merges properties, retains shared required fields and
-group-specific field dependencies, and describes the accepted input groups.
-Canonical runtime validation enforces the exact union, including exclusions
-that the advertised projection does not express. Tests validate the schemas
+The projection merges properties, retains shared required fields, and
+describes the accepted input groups. Which fields belong together is guidance
+in that description rather than a `dependentRequired` or `dependencies`
+keyword, because Gemini function declarations reject both and one unknown
+keyword fails the client's whole tool catalog. Canonical runtime validation
+enforces the exact union, including partner fields and exclusions that the
+advertised projection does not express, so a client must not rely on the
+advertised schema to reject an incomplete group. Tests validate the schemas
 after SDK conversion and check advertised validation and actual calls.
 Individual model APIs can impose additional nesting limits; complete producer
 captures can exceed ten structural levels.
@@ -689,7 +708,10 @@ reported by strict-depth MCP clients in [#2813](https://github.com/morluto/rea/i
 WorkBuddy/CodeBuddy gateways have also rejected full REA catalogs with an
 opaque `11133` error while accepting the compact profile
 ([#2999](https://github.com/morluto/rea/issues/2999)); compatibility depends on
-the selected gateway and model.
+the selected gateway and model. That report used ZCode through a gateway to a
+Kimi-K3 backend, not the WorkBuddy MCP client. Set the profile on REA's
+registration in the actual MCP host; the host's name alone does not identify
+the downstream model or its schema constraints.
 
 The compact profile renders the same tool set, output schemas, annotations,
 and canonical server-side validation. It changes only the advertised input
@@ -711,9 +733,9 @@ JSON Schema:
   in a reduced form: property names, reference-resolved types (including union
   alternatives), first-level guidance, and the root's required-field constraints,
   with nested validation detail elided. The
-  canonical schema still rejects malformed calls server-side, so callers
-  following the reduced advertisement receive the canonical typed error
-  instead of silent acceptance.
+  registered canonical schema still rejects malformed calls before invoking a
+  handler. That rejection can use the MCP SDK's input-validation diagnostic;
+  it does not guarantee the application's typed input error.
 - Schemas deeper than ten nodes use the same reduced projection, with
   root-property names, types, guidance, and required fields retained. The
   server continues to enforce the full canonical input contract.
@@ -732,6 +754,11 @@ identity. Explicit schema IDs and relative external reference bases are
 preserved. This keeps every tool and validation rule in discovery and does not
 change the SDK's catalog invalidation or availability checks.
 
+The server identity's tool-schema digest includes both the canonical output
+contracts and the selected output-schema profile. `full` and `off` therefore
+have distinct identities even though `off` keeps the same canonical output
+validation on the server.
+
 Some stdio hosts reject a complete `tools/list` frame before they load any
 tools. The opt-in `REA_MCP_OUTPUT_SCHEMA_PROFILE=off` setting omits `outputSchema`
 from the advertised catalog for hosts with a small frame limit (the measured
@@ -768,9 +795,10 @@ identities remains unknown.
   `normalization` defaults to `{ "rules": [] }`.
 
 The advertised schema includes complete nested capture fields and rejects
-incomplete pairs and structurally malformed captures. Mixed comparison families
-can pass advertised validation; the SDK rejects them against the canonical
-input schema before invoking the handler. Domain validation additionally checks
+incomplete pairs and structurally malformed captures. Mixed comparison families,
+and `normalization` without both scenario captures, can pass advertised
+validation; the SDK rejects them against the canonical input schema before
+invoking the handler. Domain validation additionally checks
 relationships such as event sequence references and retained counts; JSON
 Schema does not express those cross-field invariants.
 

docs/native-investigation.md

@@ -378,7 +378,12 @@ complete observation.
   property: `N` means `nonatomic`; a complete typed attribute string without
   `N` means the Objective-C default `atomic`;
   missing or incomplete attribute metadata leaves `atomicity` and any
-  unestablished read-only value null. `is_meta_class`, `is_root_class`, and
+  unestablished read-only value null. A type attribute whose quoted class name
+  never closes is incomplete and marks `objc_properties_categories` partial
+  with `property_attributes_incomplete`. A property record that cannot be
+  decoded, such as one with an empty or unmapped name, is reported as
+  `property_decode_failed` at its entry address while sibling properties are
+  retained. `is_meta_class`, `is_root_class`, and
   `ivar_count` are also nullable when the selected provider exposes only symbol
   names. Class implementation methods have null `is_required` and `is_optional`
   fields because those facts apply to protocol declarations; decoded protocol
@@ -481,8 +486,9 @@ user's permissions and may change files, show UI or use the network.
 
 Debugging must be allowed for the user (Developer Tools access). A
 hardened-runtime target without `com.apple.security.get-task-allow` cannot be
-debugged; `inspect_signature` reports this as the debugger-attach facet, and
-the tool fails with a `debugger-attach-denied` reason.
+debugged; `inspect_signature` reports the signature's `hardened_runtime` value
+and its `entitlements`, including `com.apple.security.get-task-allow` when it
+is present. A denied trace reports the `debugger-attach-denied` reason.
 
 ## Provider and verification boundaries
 
@@ -569,6 +575,12 @@ Exact and ambiguous value counts describe emitted root/object value instances;
 a root and an object may contain the same value. Archive digests identify the
 original bytes.
 
+Dates are reported as `{ "$plist_type": "date", "iso8601": "<timestamp>" }`. A
+date the decoder cannot represent, such as an XML date with a year beyond 9999
+or a negative year, is reported as `{ "$plist_type": "date", "iso8601": null }`
+with a counted limitation; the rest of the plist or archive is still decoded.
+`inspect_plist` and Interface Builder decoding use the same projection.
+
 ## Locate a byte signature
 
 Use `search_bytes` after opening a target with Ghidra to locate a known constant

docs/readme/README_ar.md

@@ -274,7 +274,7 @@ npx rea-agents@latest setup
 
 ## سجل النجوم
 
-🎉 **200,000 نجمة على GitHub — شكرًا لكم!**
+🎉 **250,000 نجمة على GitHub — شكرًا لكم!**
 
 شكرًا لكل من يستخدم REA، ويبلّغ عن الأخطاء، ويقترح ميزات، ويختبر إصدارات البناء، ويساهم بالإصلاحات.
 

docs/readme/README_de.md

@@ -272,7 +272,7 @@ Melde Sicherheitslücken gemäß [SECURITY.md](../../SECURITY.md).
 
 ## Sternverlauf
 
-🎉 **200.000 GitHub-Sterne – vielen Dank!**
+🎉 **250.000 GitHub-Sterne – vielen Dank!**
 
 Danke an alle, die REA nutzen, Fehler melden, Funktionswünsche äußern, Builds testen und Fehlerbehebungen beitragen.
 

docs/readme/README_el.md

@@ -285,7 +285,7 @@ npx rea-agents@latest setup
 
 ## Ιστορικό αστεριών
 
-🎉 **200.000 αστέρια στο GitHub — σας ευχαριστούμε!**
+🎉 **250.000 αστέρια στο GitHub — σας ευχαριστούμε!**
 
 Ευχαριστούμε όλους όσοι χρησιμοποιούν το REA, αναφέρουν σφάλματα, ζητούν νέες λειτουργίες, δοκιμάζουν builds και συνεισφέρουν διορθώσεις.
 

docs/readme/README_es.md

@@ -272,7 +272,7 @@ Informa de vulnerabilidades siguiendo [SECURITY.md](../../SECURITY.md).
 
 ## Historial de estrellas
 
-🎉 **200.000 estrellas en GitHub: ¡gracias!**
+🎉 **250.000 estrellas en GitHub: ¡gracias!**
 
 Gracias a quienes usan REA, informan de errores, solicitan funciones, prueban compilaciones y contribuyen con correcciones.
 

docs/readme/README_fa.md

@@ -257,7 +257,7 @@ npx rea-agents@latest setup
 
 ## تاریخچهٔ ستاره‌ها
 
-🎉 **۲۰۰٬۰۰۰ ستارهٔ GitHub — سپاسگزاریم!**
+🎉 **۲۵۰٬۰۰۰ ستارهٔ GitHub — سپاسگزاریم!**
 
 از همهٔ کسانی که از REA استفاده می‌کنند، خطاها را گزارش می‌دهند، قابلیت‌های جدید پیشنهاد می‌کنند، buildها را آزمایش می‌کنند و در بهبود پروژه مشارکت دارند سپاسگزاریم.
 

docs/readme/README_fr.md

@@ -276,7 +276,7 @@ Signalez les vulnérabilités via [SECURITY.md](../../SECURITY.md).
 
 ## Historique des étoiles
 
-🎉 **200 000 étoiles GitHub — merci !**
+🎉 **250 000 étoiles GitHub — merci !**
 
 Merci à toutes les personnes qui utilisent REA, signalent des bugs, proposent des fonctionnalités, testent les builds et contribuent des correctifs.
 

docs/readme/README_id.md

@@ -297,7 +297,7 @@ Laporkan kerentanan melalui [SECURITY.md](../../SECURITY.md).
 
 ## Riwayat Bintang
 
-🎉 **200.000 bintang GitHub, terima kasih!**
+🎉 **250.000 bintang GitHub, terima kasih!**
 
 Terima kasih kepada semua orang yang menggunakan REA, melaporkan bug, meminta fitur, menguji build, dan berkontribusi dalam perbaikan.
 

docs/readme/README_it.md

@@ -276,7 +276,7 @@ Segnala vulnerabilità tramite [SECURITY.md](../../SECURITY.md).
 
 ## Cronologia delle stelle
 
-🎉 **200.000 stelle su GitHub — grazie!**
+🎉 **250.000 stelle su GitHub — grazie!**
 
 Grazie a tutti coloro che usano REA, segnalano bug, propongono funzionalità, testano le build e contribuiscono con correzioni.
 

docs/readme/README_ja.md

@@ -274,7 +274,7 @@ npx rea-agents@latest setup
 
 ## スター履歴
 
-🎉 **GitHub スター 200,000 件、ありがとうございます！**
+🎉 **GitHub スター 250,000 件、ありがとうございます！**
 
 REA の利用、バグ報告、機能の要望、ビルドのテスト、修正への貢献に感謝します。
 

docs/readme/README_ko.md

@@ -283,7 +283,7 @@ npx rea-agents@latest setup
 
 ## 스타 기록
 
-🎉 **GitHub 스타 200,000개, 감사합니다!**
+🎉 **GitHub 스타 250,000개, 감사합니다!**
 
 REA를 사용하고, 버그를 보고하고, 기능을 요청하고, 빌드를 테스트하고, 수정에 기여해 주신 모든 분께 감사드립니다.
 

docs/readme/README_pl.md

@@ -272,7 +272,7 @@ Zgłaszaj podatności zgodnie z [SECURITY.md](../../SECURITY.md).
 
 ## Historia gwiazdek
 
-🎉 **200 000 gwiazdek na GitHubie — dziękujemy!**
+🎉 **250 000 gwiazdek na GitHubie — dziękujemy!**
 
 Dziękujemy wszystkim, którzy używają REA, zgłaszają błędy, proponują nowe funkcje, testują kompilacje i przesyłają poprawki.
 

docs/readme/README_pt-BR.md

@@ -272,7 +272,7 @@ Relate vulnerabilidades conforme [SECURITY.md](../../SECURITY.md).
 
 ## Histórico de estrelas
 
-🎉 **200.000 estrelas no GitHub — muito obrigado!**
+🎉 **250.000 estrelas no GitHub — muito obrigado!**
 
 Obrigado a todos que usam o REA, relatam bugs, sugerem funcionalidades, testam builds e contribuem com correções.
 

docs/readme/README_ru.md

@@ -272,7 +272,7 @@ npx rea-agents@latest setup
 
 ## История звёзд
 
-🎉 **200 000 звёзд на GitHub — спасибо!**
+🎉 **250 000 звёзд на GitHub — спасибо!**
 
 Спасибо всем, кто использует REA, сообщает об ошибках, предлагает новые функции, тестирует сборки и помогает с исправлениями.
 

docs/readme/README_th.md

@@ -272,7 +272,7 @@ npx rea-agents@latest setup
 
 ## ประวัติดาว
 
-🎉 **200,000 ดาวบน GitHub — ขอบคุณทุกคน!**
+🎉 **250,000 ดาวบน GitHub — ขอบคุณทุกคน!**
 
 ขอบคุณทุกคนที่ใช้ REA รายงานบั๊ก เสนอฟีเจอร์ ทดสอบบิลด์ และช่วยแก้ไข
 

docs/readme/README_tr.md

@@ -277,7 +277,7 @@ Güvenlik açıklarını [SECURITY.md](../../SECURITY.md) üzerinden bildirin.
 
 ## Yıldız geçmişi
 
-🎉 **GitHub'da 200.000 yıldız — teşekkürler!**
+🎉 **GitHub'da 250.000 yıldız — teşekkürler!**
 
 REA'yı kullanan, hata bildiren, özellik talep eden, derlemeleri test eden ve düzeltmelere katkıda bulunan herkese teşekkürler.
 

docs/readme/README_uk.md

@@ -272,7 +272,7 @@ npx rea-agents@latest setup
 
 ## Історія зірок
 
-🎉 **200 000 зірок на GitHub — дякуємо!**
+🎉 **250 000 зірок на GitHub — дякуємо!**
 
 Дякуємо всім, хто використовує REA, повідомляє про помилки, пропонує нові функції, тестує збірки й допомагає з виправленнями.
 

docs/readme/README_vi.md

@@ -272,7 +272,7 @@ Báo cáo lỗ hổng theo [SECURITY.md](../../SECURITY.md).
 
 ## Lịch sử sao
 
-🎉 **200.000 sao trên GitHub — xin cảm ơn!**
+🎉 **250.000 sao trên GitHub — xin cảm ơn!**
 
 Cảm ơn mọi người đã sử dụng REA, báo lỗi, đề xuất tính năng, kiểm thử bản dựng và đóng góp bản sửa lỗi.
 

docs/readme/README_zh-TW.md

@@ -278,7 +278,7 @@ npx rea-agents@latest setup
 
 ## Star 歷史
 
-🎉 **GitHub Star 達到 200,000 個，感謝大家！**
+🎉 **GitHub Star 達到 250,000 個，感謝大家！**
 
 感謝每一位使用 REA、報告錯誤、提出功能需求、測試建置和貢獻修復的朋友。
 

docs/readme/README_zh.md

@@ -278,7 +278,7 @@ npx rea-agents@latest setup
 
 ## Star 历史
 
-🎉 **GitHub Star 达到 200,000 个，感谢大家！**
+🎉 **GitHub Star 达到 250,000 个，感谢大家！**
 
 感谢每一位使用 REA、报告错误、提出功能需求、测试构建和贡献修复的朋友。
 

docs/testing.md

@@ -98,6 +98,18 @@ isolated filesystem boundary tests because it loads both source and compiled
 module graphs. Adapter, other boundary, acceptance, and process-global projects
 retain per-file isolation. See `vitest.config.ts` for current settings.
 
+Inject producer faults through scoped dependencies in shared workers: a module
+mock cannot replace a consumer that was imported earlier. Check these tests
+with the consumer preloaded as well as alone. For simulated protocol traces,
+use an acknowledged event barrier before asserting complete observations;
+a zero-duration capture does not guarantee queued events have arrived.
+
+Check advertised schemas after SDK conversion for every supported schema
+profile. Meta-schema validity does not establish client acceptance or useful
+rendered declarations. Keep canonical rejection checks separate from the
+intentionally reduced compact advertisement, and probe actual client renderers
+when changing schema representations.
+
 The reused CLI boundary workers cap V8 old space at 384 MiB, based on their
 profiled allocation and cleanup behavior. This encourages collection between
 files; spawned REA processes retain their own heap settings.

docs/tool-design.md

@@ -87,6 +87,12 @@ malformed, unavailable, cancellation, and partial outcomes through affected
 consumers. Real-engine claims require the matching provider lane; see
 [testing.md](testing.md).
 
+For fixed tuples in public contracts, use `src/domain/fixedTupleSchema.ts`:
+Zod's tuple projection omits implicit length bounds and Draft 2020-12 `items`,
+which some clients require even alongside `prefixItems`. Tuples with a rest
+schema also need explicit minimum lengths. Check parser/schema parity for empty,
+short, overlong, and malformed values in both input and output projections.
+
 When discoverability changes, evaluate representative broad, direct, and negative
 tasks on the intended host. Record host/model/catalog, selected tools, arguments,
 errors, retries, and task outcomes. Schema validity and keyword heuristics do not

docs/web-network-captures.md

@@ -117,7 +117,11 @@ gRPC message bytes. Missing members remain null, distinct from empty bodies.
 See Telerik's [SAZ format reference](https://www.telerik.com/fiddler/fiddler-everywhere/documentation/knowledge-base/fiddler-archives).
 
 The existing ZIP reader verifies read members; expanded archive contents are
-limited to 32 MiB in addition to the input/output budgets below. Encrypted
+limited to 32 MiB in addition to the input/output budgets below. Exceeding this
+expanded-byte budget returns `invalid_request` / `invalid_input` through both
+CLI and MCP, with an `out_of_range` issue at
+`capture_path` and `maximum_expanded_capture_bytes`. Selecting records does not
+bypass the complete-archive admission check or return partial Evidence. Encrypted
 archives and unrecognized raw session members are unsupported. SAZ requires an
 enforced decoder memory bound: Node's fixed heap, or the Windows Job Object
 commit limit. Bun on hosts without that Job Object limit returns capability

experiments/analysis-efficiency/README.md

@@ -1,5 +1,11 @@
 # Analysis efficiency measurements
 
+For genuine Fiddler captures, see the [SAZ selection checkpoint](saz-selection-workflows.md).
+`measure.mjs --saz-selection baseline|focused|retained` compares default complete
+delivery (including retained-reference recovery), caller-selected ordinals, and
+summary/preview/complete/repeated pages. Each failed RPC remains in task costs.
+The body question uses stored bytes, not decoded HTTP application payloads.
+
 Tracks the experiments for [#2016](https://github.com/morluto/rea/issues/2016).
 The production CLI/MCP contracts and defaults remain the compatibility baseline.
 Downloaded artifacts, runtime copies, exported Evidence, transcripts and detailed

experiments/analysis-efficiency/agent-trial.py

@@ -181,6 +181,7 @@ def main():
                  'inspect_android_class', 'inspect_android_method',
                  'inspect_firmware_regions', 'inspect_managed_artifact',
                  'inspect_sqlite_database', 'inspect_pcap_capture',
+                 'inspect_web_network_capture',
                  'inspect_packet_capture', 'inspect_packet_fields',
                  'decode_signal_recording', 'inspect_signal_decoders',
                  'inspect_managed_members', 'inspect_managed_native_boundaries',

experiments/analysis-efficiency/corpus.json

@@ -373,6 +373,66 @@
         "reference": "https://raw.githubusercontent.com/sigrokproject/sigrok-dumps/0ad13477abc959d37fc9a5acbd23901c371c9c76/i2c/wii_nunchuk/README",
         "license_scope": "Repository README declares dumps public domain unless otherwise noted; selected READMEs state no exception; decoder software licensing is separate."
       }
+    },
+    {
+      "id": "xbox-clear_nomp.saz",
+      "file": "clear_nomp.saz",
+      "bytes": 43515,
+      "sha256": "3e808663a99cbb9160186f7419633248200fbc6863d2c897f5d4d9972dde0375",
+      "provenance": {
+        "project": "Microsoft Xbox Live Trace Analyzer",
+        "version": "ad880ff00b26c4cae5e3c8864d3df438a03890ba",
+        "url": "https://raw.githubusercontent.com/microsoft/xbox-live-trace-analyzer/ad880ff00b26c4cae5e3c8864d3df438a03890ba/Source/TestData/CertWarnings/clear_nomp.saz",
+        "git_blob": "2f1ba66dacb8f8bac290cab59dff8fe750a46ce7",
+        "license": "MIT; upstream LICENSE.md",
+        "producer_archive_comment": "Fiddler (v4.5.1.2) Session Archive. See http://fiddler2.com"
+      },
+      "saz_question": {
+        "record_ordinals": [12, 0, 6],
+        "total_records": 13,
+        "members": 42,
+        "expanded_bytes": 62609
+      }
+    },
+    {
+      "id": "xbox-clear_mp_cs10.saz",
+      "file": "clear_mp_cs10.saz",
+      "bytes": 5223519,
+      "sha256": "4021b201848e4feea3402c380fda5d2df43fd0bf1c72ea550e60f2da3b3d5b44",
+      "provenance": {
+        "project": "Microsoft Xbox Live Trace Analyzer",
+        "version": "ad880ff00b26c4cae5e3c8864d3df438a03890ba",
+        "url": "https://raw.githubusercontent.com/microsoft/xbox-live-trace-analyzer/ad880ff00b26c4cae5e3c8864d3df438a03890ba/Source/TestData/CertWarnings/clear_mp_cs10.saz",
+        "git_blob": "dac534d625594c7a1d57b0b12ed12ea9c6cb6a5b",
+        "license": "MIT; upstream LICENSE.md",
+        "producer_archive_comment": "Fiddler (v4.6.2.0) Session Archive. See https://fiddler2.com"
+      },
+      "saz_question": {
+        "record_ordinals": [1310, 0, 655],
+        "total_records": 1311,
+        "members": 3939,
+        "expanded_bytes": 9223820
+      }
+    },
+    {
+      "id": "xbox-x49_mp_cs21.saz",
+      "file": "x49_mp_cs21.saz",
+      "bytes": 572224,
+      "sha256": "97cdca8b0608e3f11995fd64b64390b9fd8d8e3efcac89f517f24b44968047f6",
+      "provenance": {
+        "project": "Microsoft Xbox Live Trace Analyzer",
+        "version": "ad880ff00b26c4cae5e3c8864d3df438a03890ba",
+        "url": "https://raw.githubusercontent.com/microsoft/xbox-live-trace-analyzer/ad880ff00b26c4cae5e3c8864d3df438a03890ba/Source/TestData/CertWarnings/x49_mp_cs21.saz",
+        "git_blob": "e7794c44892ca4fa1729fa012e698e8284337f15",
+        "license": "MIT; upstream LICENSE.md",
+        "producer_archive_comment": "Fiddler (v4.6.2.26757) Session Archive. See https://fiddler2.com"
+      },
+      "saz_question": {
+        "record_ordinals": [102, 0, 51],
+        "total_records": 103,
+        "members": 315,
+        "expanded_bytes": 1301927
+      }
     }
   ]
 }

experiments/analysis-efficiency/measure.mjs

@@ -14,6 +14,7 @@ import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
 import { Ajv2020 } from "ajv/dist/2020.js";
 import { PACKET_ARTIFACTS, measurePacketEntry } from "./packet-entry.mjs";
 import { SIGNAL_ARTIFACTS, measureSignalEntry } from "./signal-entry.mjs";
+import { SAZ_ARTIFACTS, measureSazEntry } from "./saz-entry.mjs";
 
 const root = resolve(dirname(fileURLToPath(import.meta.url)), "../..");
 const { values } = parseArgs({
@@ -33,6 +34,7 @@ const { values } = parseArgs({
     "sqlite-selection": { type: "string", default: "all" },
     "packet-selection": { type: "string", default: "baseline" },
     "signal-selection": { type: "string", default: "baseline" },
+    "saz-selection": { type: "string", default: "baseline" },
   },
 });
 if (
@@ -43,6 +45,11 @@ if (
   !isAbsolute(values.path) ||
   !isAbsolute(values.output) ||
   !["complete", "summary"].includes(values.delivery) ||
+  !["baseline", "focused", "retained"].includes(values["saz-selection"]) ||
+  (!Object.hasOwn(SAZ_ARTIFACTS, values.artifact) &&
+    values["saz-selection"] !== "baseline") ||
+  (Object.hasOwn(SAZ_ARTIFACTS, values.artifact) &&
+    values.delivery !== "complete") ||
   !["baseline", "focused", "discovery", "pages"].includes(
     values["signal-selection"],
   ) ||
@@ -174,6 +181,12 @@ const fingerprints = {
   signal_oracle_source_sha256: await fileSha(
     fileURLToPath(new URL("./signal-oracle.py", import.meta.url)),
   ),
+  saz_workflow_source_sha256: await fileSha(
+    fileURLToPath(new URL("./saz-entry.mjs", import.meta.url)),
+  ),
+  saz_oracle_source_sha256: await fileSha(
+    fileURLToPath(new URL("./saz-oracle.py", import.meta.url)),
+  ),
   runtime_files_sha256: sha(JSON.stringify(runtimeFiles)),
   runtime_dispatcher_sha256: await fileSha(
     resolve(values["runtime-root"], "scripts/rea.mjs"),
@@ -754,6 +767,12 @@ const inspectPcapTarget = async () => {
 };
 
 const workflows = {
+  ...Object.fromEntries(
+    Object.keys(SAZ_ARTIFACTS).map((artifact) => [
+      artifact,
+      () => measureSazEntry({ values, call, inputSha }),
+    ]),
+  ),
   ...Object.fromEntries(
     Object.keys(SIGNAL_ARTIFACTS).map((artifact) => [
       artifact,
@@ -1188,6 +1207,9 @@ const report = {
   artifact_sha256: inputSha,
   artifact_sha256_after: finalInputSha,
   delivery: values.delivery,
+  saz_selection: Object.hasOwn(SAZ_ARTIFACTS, values.artifact)
+    ? values["saz-selection"]
+    : null,
   sdk_schema_mode: values.schemas,
   retained_verification: values["verify-retained"],
   native_question: NATIVE_ARTIFACTS.includes(values.artifact)

experiments/analysis-efficiency/review-saz.mjs

@@ -0,0 +1,266 @@
+import assert from "node:assert/strict";
+import { readFile, writeFile } from "node:fs/promises";
+import { resolve } from "node:path";
+import { pathToFileURL } from "node:url";
+import { isDeepStrictEqual, parseArgs } from "node:util";
+import {
+  sazRequest,
+  verifySazRecords,
+  verifySazOracleIdentity,
+  sazEvidenceParameters,
+} from "./saz-entry.mjs";
+
+const { values } = parseArgs({
+  options: {
+    trial: { type: "string" },
+    oracle: { type: "string" },
+    "runtime-root": { type: "string" },
+    artifact: { type: "string" },
+    selection: { type: "string" },
+  },
+});
+if (
+  !values.trial ||
+  !values.oracle ||
+  !values["runtime-root"] ||
+  !["baseline", "focused"].includes(values.selection)
+)
+  throw new Error(
+    "Use --trial DIR --oracle JSON --runtime-root DIR --artifact ID --selection baseline|focused",
+  );
+const load = async (file) =>
+  JSON.parse(await readFile(resolve(values.trial, file), "utf8"));
+const settings = await load("trial-settings.json"),
+  summary = await load("summary.json");
+const oracle = JSON.parse(await readFile(values.oracle, "utf8"));
+verifySazOracleIdentity(values.artifact, oracle);
+const runtime = (file) =>
+  pathToFileURL(resolve(values["runtime-root"], file)).href;
+const { parseEvidence } = await import(runtime("dist/domain/evidence.js"));
+const { inspectWebNetworkCaptureInputSchema, webNetworkCaptureSchema } =
+  await import(runtime("dist/domain/webNetworkCapture.js"));
+const { inspectAnalysisViewInputSchema, analysisViewResultSchema } =
+  await import(runtime("dist/domain/analysisView/analysisView.js"));
+const calls = (await readFile(resolve(values.trial, "events.jsonl"), "utf8"))
+  .split("\n")
+  .filter(Boolean)
+  .map(JSON.parse)
+  .filter(
+    (event) =>
+      event.type === "item.completed" && event.item?.type === "mcp_tool_call",
+  )
+  .map((event) => event.item);
+const planned = inspectWebNetworkCaptureInputSchema.parse(
+  sazRequest(
+    values.artifact,
+    settings.selected_artifact_path,
+    values.selection,
+  ),
+);
+const reasons = [],
+  workflowReasons = [],
+  authenticated = new Map(),
+  parents = new Set(),
+  conflicts = new Set();
+let transportRecovery = false;
+for (const call of calls) {
+  try {
+    assert.equal(call.server, "rea");
+    assert.equal(call.status, "completed");
+    assert.ok(!call.error, "RPC did not provide observations");
+    const text = call.result.content.filter((item) => item.type === "text");
+    assert.equal(text.length, 1, "Ambiguous Evidence text carrier");
+    let evidence = JSON.parse(text[0].text);
+    if (call.result.isError) {
+      const limits = evidence.error?.details?.reported_limits;
+      assert.equal(call.tool, "inspect_web_network_capture");
+      assert.equal(evidence.error?.code, "resource_constraint");
+      assert.equal(
+        evidence.error.details.operation,
+        "inspect_web_network_capture",
+      );
+      assert.equal(limits?.boundary, "mcp-response");
+      const input = inspectWebNetworkCaptureInputSchema.parse(call.arguments);
+      assert.deepEqual(input, planned);
+      const source = inspectAnalysisViewInputSchema.parse({
+        source: limits.evidence_reference,
+        view: { kind: "summary" },
+      }).source;
+      assert.equal(source.kind, "retained-evidence");
+      parents.add(source.evidence_id);
+      transportRecovery = true;
+      continue;
+    }
+    if (call.result.structuredContent !== undefined)
+      assert.deepEqual(evidence, call.result.structuredContent);
+    evidence = parseEvidence(evidence);
+    assert.equal(evidence.subject.local_path, settings.selected_artifact_path);
+    assert.equal(evidence.subject.digest.sha256, oracle.sha256);
+    let records;
+    if (call.tool === "inspect_web_network_capture") {
+      const { detail, ...arguments_ } = call.arguments;
+      const input = inspectWebNetworkCaptureInputSchema.parse(arguments_);
+      assert.equal(input.capture_path, settings.selected_artifact_path);
+      assert.equal(input.format, "saz");
+      if (!isDeepStrictEqual(input, planned))
+        workflowReasons.push(
+          "Inspection selection differs from planned arguments",
+        );
+      if (detail === "summary") {
+        workflowReasons.push(
+          "Summary delivery differs from the fixed complete-delivery plan",
+        );
+        assert.equal(evidence.operation, "inspect_analysis_view");
+        const report = analysisViewResultSchema.parse(
+          evidence.normalized_result,
+        );
+        assert.equal(report.kind, "summary");
+        assert.equal(report.parent_operation, "inspect_web_network_capture");
+        assert.equal(report.summary.format, "saz");
+        assert.equal(report.summary.total_records, oracle.total_records);
+        assert.deepEqual(evidence.parameters, {
+          parent_evidence_id: report.parent_evidence_id,
+          view: { kind: "summary" },
+        });
+        parents.add(report.parent_evidence_id);
+        authenticated.set(evidence.evidence_id, []);
+        continue;
+      }
+      assert.ok(detail === undefined || detail === "complete");
+      assert.equal(evidence.operation, "inspect_web_network_capture");
+      assert.deepEqual(
+        evidence.parameters,
+        sazEvidenceParameters(input, oracle.total_records),
+      );
+      const report = webNetworkCaptureSchema.parse(evidence.normalized_result);
+      assert.equal(report.total_records, oracle.total_records);
+      assert.equal(report.format, "saz");
+      records = report.records;
+      parents.add(evidence.evidence_id);
+    } else {
+      assert.equal(call.tool, "inspect_analysis_view");
+      const input = inspectAnalysisViewInputSchema.parse(call.arguments);
+      assert.equal(input.source.kind, "retained-evidence");
+      assert.ok(
+        parents.has(input.source.evidence_id),
+        "Unknown retained parent",
+      );
+      assert.equal(evidence.operation, "inspect_analysis_view");
+      assert.deepEqual(evidence.parameters, {
+        parent_evidence_id: input.source.evidence_id,
+        view: input.view,
+      });
+      const page = analysisViewResultSchema.parse(evidence.normalized_result);
+      assert.deepEqual(page.view, input.view);
+      assert.equal(page.parent_evidence_id, input.source.evidence_id);
+      assert.equal(page.parent_operation, "inspect_web_network_capture");
+      if (page.kind === "summary") {
+        assert.equal(page.summary.total_records, oracle.total_records);
+        assert.equal(page.summary.format, "saz");
+        records = [];
+      } else {
+        assert.equal(page.kind, "capture");
+        records = page.view.detail === "complete" ? page.items : [];
+      }
+    }
+    const prior = authenticated.get(evidence.evidence_id);
+    if (prior && !isDeepStrictEqual(prior, records)) {
+      conflicts.add(evidence.evidence_id);
+      authenticated.delete(evidence.evidence_id);
+      parents.delete(evidence.evidence_id);
+      throw new Error("Conflicting observations share an Evidence ID");
+    }
+    if (!conflicts.has(evidence.evidence_id))
+      authenticated.set(evidence.evidence_id, records);
+  } catch (cause) {
+    workflowReasons.push(String(cause));
+  }
+}
+if (
+  calls[0]?.tool !== "inspect_web_network_capture" ||
+  calls.filter((call) => call.tool === "inspect_web_network_capture").length !==
+    1 ||
+  (!transportRecovery && calls.length !== 1)
+)
+  workflowReasons.push(
+    "Completed call sequence differs from the fixed initial call/recovery plan",
+  );
+try {
+  const raw = (await readFile(resolve(values.trial, "answer.txt"), "utf8"))
+    .trim()
+    .replace(/^```(?:json)?\s*|\s*```$/g, "");
+  const answer = JSON.parse(raw);
+  assert.ok(
+    Array.isArray(answer.evidence_ids) && answer.evidence_ids.length > 0,
+    "Missing Evidence citations",
+  );
+  const observed = new Map();
+  for (const id of answer.evidence_ids) {
+    assert.ok(authenticated.has(id), "Unknown/failed cited Evidence");
+    for (const row of authenticated.get(id)) {
+      const prior = observed.get(row.ordinal);
+      assert.ok(
+        !prior || isDeepStrictEqual(prior, row),
+        "Conflicting cited record identities",
+      );
+      observed.set(row.ordinal, row);
+    }
+  }
+  const facts = verifySazRecords(
+    oracle.record_ordinals.map((ordinal) => observed.get(ordinal)),
+    oracle,
+  );
+  assert.equal(answer.records.length, facts.length, "Answer coverage differs");
+  assert.deepEqual(
+    [...answer.records].sort((a, b) => a.ordinal - b.ordinal),
+    [...facts].sort((a, b) => a.ordinal - b.ordinal),
+    "Answer facts differ from cited stored bytes",
+  );
+  if (
+    !isDeepStrictEqual(
+      answer.records.map((row) => row.ordinal),
+      oracle.record_ordinals,
+    )
+  )
+    workflowReasons.push("Final answer changes caller-selected order");
+  if (
+    !Array.isArray(answer.limitations) ||
+    answer.limitations.length === 0 ||
+    answer.limitations.some(
+      (value) => typeof value !== "string" || !value.trim(),
+    )
+  )
+    workflowReasons.push("Final answer omits limitations");
+  assert.equal(settings.selected_artifact_sha256_before, oracle.sha256);
+  assert.equal(summary.selected_artifact_sha256_after, oracle.sha256);
+  if (
+    summary.exit_code !== 0 ||
+    summary.timeout ||
+    summary.selected_artifact_bytes_unchanged !== true
+  )
+    workflowReasons.push(
+      "Trial did not complete with original bytes unchanged",
+    );
+} catch (cause) {
+  reasons.push(String(cause));
+}
+const report = {
+  factual_correctness: reasons.length ? "failed" : "passed",
+  workflow_correctness:
+    reasons.length || workflowReasons.length ? "failed" : "passed",
+  reasons,
+  workflow_reasons: workflowReasons,
+  completed_calls: calls.length,
+  wall_seconds: summary.wall_seconds,
+  usage: summary.usage,
+  selected_artifact_sha256: oracle.sha256,
+  scope:
+    "Authenticated cited stored HTTP records and independent raw-byte checks; no live attribution, body decompression or certificate interpretation",
+  limitation_prose: "Presence checked; wording requires transcript review",
+};
+await writeFile(
+  resolve(values.trial, "saz-review.json"),
+  JSON.stringify(report, null, 2) + "\n",
+);
+console.log(JSON.stringify(report));
+if (report.workflow_correctness !== "passed") process.exitCode = 1;

experiments/analysis-efficiency/saz-entry.mjs

@@ -0,0 +1,326 @@
+import assert from "node:assert/strict";
+import { execFile } from "node:child_process";
+import { readFile } from "node:fs/promises";
+import { resolve } from "node:path";
+import { fileURLToPath, pathToFileURL } from "node:url";
+import { promisify } from "node:util";
+
+const corpus = JSON.parse(
+  await readFile(new URL("./corpus.json", import.meta.url), "utf8"),
+);
+export const SAZ_ARTIFACTS = Object.fromEntries(
+  corpus.artifacts
+    .filter((entry) => entry.saz_question)
+    .map((entry) => [entry.id, entry]),
+);
+
+export function sazRequest(artifact, path, selection) {
+  const question = Object.hasOwn(SAZ_ARTIFACTS, artifact)
+    ? SAZ_ARTIFACTS[artifact].saz_question
+    : undefined;
+  if (!question) throw new Error("Select a pinned SAZ question");
+  return {
+    capture_path: path,
+    format: "saz",
+    ...(selection === "baseline"
+      ? {}
+      : { record_ordinals: question.record_ordinals }),
+    sensitive_values: [],
+  };
+}
+
+export function verifySazOracleIdentity(artifact, oracle) {
+  assert.ok(
+    Object.hasOwn(SAZ_ARTIFACTS, artifact),
+    "Unknown pinned SAZ question",
+  );
+  const entry = SAZ_ARTIFACTS[artifact];
+  assert.equal(oracle.kind, "saz");
+  assert.equal(oracle.artifact, artifact);
+  assert.equal(oracle.sha256, entry.sha256, "Wrong immutable corpus");
+  assert.equal(oracle.total_records, entry.saz_question.total_records);
+  assert.deepEqual(oracle.record_ordinals, entry.saz_question.record_ordinals);
+}
+
+export function sazEvidenceParameters(input, totalRecords) {
+  return {
+    capture_path: input.capture_path,
+    format: input.format,
+    record_ordinals:
+      input.record_ordinals ??
+      Array.from({ length: totalRecords }, (_, index) => index),
+    sensitive_value_count: input.sensitive_values.length,
+  };
+}
+
+export function verifySazRecords(records, oracle) {
+  assert.equal(oracle.kind, "saz", "Wrong independent question");
+  assert.equal(
+    records.length,
+    oracle.records.length,
+    "Selected coverage differs",
+  );
+  return records.map((record, index) => {
+    const raw = oracle.records[index];
+    assert.equal(record.ordinal, raw.ordinal, "Caller order/ordinal differs");
+    assert.equal(
+      record.reported.session_id,
+      raw.session_id,
+      "Padded source identity differs",
+    );
+    assert.deepEqual(
+      record.location,
+      { kind: "archive-members", members: raw.members },
+      "Original member coordinates differ",
+    );
+    for (const side of ["request", "response"]) {
+      const expected = raw[side],
+        reported = record.reported[side];
+      if (expected === null) {
+        assert.equal(reported, null, "Absent message invented");
+        assert.ok(
+          !record.binary_fields.some(
+            (field) => field.pointer === `/${side}/body`,
+          ),
+          "Absent body invented",
+        );
+        continue;
+      }
+      assert.equal(reported.member, expected.member);
+      for (const field of side === "request"
+        ? ["method", "url", "http_version"]
+        : ["status", "reason", "http_version"])
+        assert.equal(
+          reported[field],
+          expected[field],
+          "Stored HTTP start line differs",
+        );
+      const contentTypes = (headers) =>
+        headers.filter(
+          (header) => header.name.toLowerCase() === "content-type",
+        );
+      assert.deepEqual(
+        contentTypes(reported.headers),
+        contentTypes(expected.headers),
+        "Content-Type order/spelling/values differ",
+      );
+      const bodies = record.binary_fields.filter(
+        (field) => field.pointer === `/${side}/body`,
+      );
+      assert.equal(bodies.length, 1, "Stored body representation differs");
+      const body = bodies[0];
+      assert.equal(body.state, "retained", "Stored body unexpectedly unknown");
+      assert.equal(
+        body.representation,
+        "producer-bytes",
+        "Encoded source representation differs",
+      );
+      for (const field of ["bytes", "sha256", "content_base64"])
+        assert.equal(
+          body[field],
+          expected.body[field],
+          "Stored encoded body bytes differ",
+        );
+      assert.equal(
+        reported.body,
+        expected.body.text,
+        "UTF-8 projection or empty body differs",
+      );
+    }
+    const response = raw.response;
+    return {
+      ordinal: record.ordinal,
+      session_id: raw.session_id,
+      members: raw.members,
+      request:
+        raw.request === null
+          ? null
+          : { method: raw.request.method, url: raw.request.url },
+      response:
+        response === null
+          ? null
+          : {
+              status: response.status,
+              http_version: response.http_version,
+              content_types: response.headers.filter(
+                (header) => header.name.toLowerCase() === "content-type",
+              ),
+              body: {
+                bytes: response.body.bytes,
+                sha256: response.body.sha256,
+              },
+            },
+    };
+  });
+}
+
+export async function measureSazEntry({ values, call, inputSha }) {
+  const runtime = (file) =>
+    pathToFileURL(resolve(values["runtime-root"], file)).href;
+  const { parseEvidence } = await import(runtime("dist/domain/evidence.js"));
+  const { inspectWebNetworkCaptureInputSchema, webNetworkCaptureSchema } =
+    await import(runtime("dist/domain/webNetworkCapture.js"));
+  const { inspectAnalysisViewInputSchema, analysisViewResultSchema } =
+    await import(runtime("dist/domain/analysisView/analysisView.js"));
+  const read = await promisify(execFile)(
+    "python3",
+    [
+      fileURLToPath(new URL("./saz-oracle.py", import.meta.url)),
+      values.artifact,
+      values.path,
+    ],
+    { maxBuffer: 2 * 1024 * 1024, timeout: 30_000 },
+  );
+  const oracle = JSON.parse(read.stdout);
+  verifySazOracleIdentity(values.artifact, oracle);
+  assert.equal(oracle.sha256, inputSha);
+  const input = inspectWebNetworkCaptureInputSchema.parse(
+    sazRequest(values.artifact, values.path, values["saz-selection"]),
+  );
+  const observations = [];
+  const checked = (evidence, schema) => {
+    evidence = parseEvidence(evidence);
+    assert.equal(evidence.subject.local_path, values.path);
+    assert.equal(evidence.subject.digest.sha256, inputSha);
+    const report = schema.parse(evidence.normalized_result);
+    observations.push({
+      evidence_id: evidence.evidence_id,
+      operation: evidence.operation,
+      parameters: evidence.parameters,
+      limitations: evidence.limitations,
+      coverage: report.coverage ?? {
+        total_records: report.total_records,
+        retained_records: report.records.length,
+      },
+    });
+    return report;
+  };
+  const readView = async (source, view) => {
+    const request = inspectAnalysisViewInputSchema.parse({ source, view });
+    const evidence = await call("inspect_analysis_view", request);
+    assert.equal(evidence.operation, "inspect_analysis_view");
+    assert.deepEqual(evidence.parameters, {
+      parent_evidence_id: source.evidence_id,
+      view: request.view,
+    });
+    const result = checked(evidence, analysisViewResultSchema);
+    assert.equal(result.kind, request.view.kind);
+    assert.equal(result.parent_evidence_id, source.evidence_id);
+    assert.equal(result.parent_operation, "inspect_web_network_capture");
+    assert.deepEqual(result.view, request.view);
+    return result;
+  };
+  let records, decoder;
+  if (values["saz-selection"] !== "retained") {
+    let evidence;
+    try {
+      evidence = await call("inspect_web_network_capture", input);
+    } catch (failure) {
+      const error = failure.value?.error;
+      const limits = error?.details?.reported_limits;
+      if (
+        values["saz-selection"] !== "baseline" ||
+        error?.code !== "resource_constraint" ||
+        error.details.operation !== "inspect_web_network_capture" ||
+        limits?.boundary !== "mcp-response" ||
+        !limits.evidence_reference
+      )
+        throw failure;
+      // Recover only the exact retained reference returned by this failed call.
+      // The failed RPC remains in the shared measurement log and task costs.
+      const source = limits.evidence_reference;
+      assert.equal(source.kind, "retained-evidence");
+      const summary = await readView(source, { kind: "summary" });
+      assert.equal(summary.summary.format, "saz");
+      assert.equal(summary.summary.total_records, oracle.total_records);
+      assert.equal(summary.summary.retained_records, oracle.total_records);
+      decoder = summary.summary.decoder;
+      records = [];
+      for (const ordinal of oracle.record_ordinals) {
+        const page = await readView(source, {
+          kind: "capture",
+          detail: "complete",
+          offset: ordinal,
+          limit: 1,
+        });
+        assert.equal(page.coverage.total, oracle.total_records);
+        assert.equal(page.items.length, 1);
+        records.push(page.items[0]);
+      }
+    }
+    if (evidence !== undefined) {
+      assert.equal(evidence.operation, "inspect_web_network_capture");
+      assert.deepEqual(
+        evidence.parameters,
+        sazEvidenceParameters(input, oracle.total_records),
+      );
+      const report = checked(evidence, webNetworkCaptureSchema);
+      assert.equal(report.format, "saz");
+      assert.equal(report.total_records, oracle.total_records);
+      assert.equal(
+        report.records.length,
+        values["saz-selection"] === "baseline"
+          ? oracle.total_records
+          : oracle.records.length,
+      );
+      if (values["saz-selection"] === "baseline")
+        assert.deepEqual(
+          report.records.map((record) => record.ordinal),
+          Array.from({ length: oracle.total_records }, (_, index) => index),
+        );
+      records =
+        values["saz-selection"] === "baseline"
+          ? oracle.record_ordinals.map((ordinal) =>
+              report.records.find((record) => record.ordinal === ordinal),
+            )
+          : report.records;
+      decoder = report.decoder;
+    }
+  } else {
+    const initial = await call("inspect_web_network_capture", {
+      ...input,
+      detail: "summary",
+    });
+    const summary = checked(initial, analysisViewResultSchema);
+    assert.equal(summary.kind, "summary");
+    assert.equal(initial.operation, "inspect_analysis_view");
+    assert.deepEqual(initial.parameters, {
+      parent_evidence_id: summary.parent_evidence_id,
+      view: { kind: "summary" },
+    });
+    assert.equal(summary.summary.format, "saz");
+    assert.equal(summary.parent_operation, "inspect_web_network_capture");
+    assert.equal(summary.summary.total_records, oracle.total_records);
+    assert.equal(summary.summary.retained_records, oracle.records.length);
+    decoder = summary.summary.decoder;
+    const source = {
+      kind: "retained-evidence",
+      evidence_id: summary.parent_evidence_id,
+    };
+    const page = async (detail, offset, limit) => {
+      return readView(source, { kind: "capture", detail, offset, limit });
+    };
+    const preview = await page("preview", 0, 3);
+    assert.deepEqual(
+      preview.items.map((record) => record.ordinal),
+      oracle.record_ordinals,
+    );
+    const first = await page("complete", 0, 2),
+      next = await page("complete", 2, 2),
+      repeated = await page("complete", 0, 2);
+    assert.deepEqual(first.items, repeated.items);
+    assert.deepEqual(first.coverage, repeated.coverage);
+    assert.equal(first.coverage.next_offset, 2);
+    assert.equal(first.coverage.exhausted, false);
+    assert.equal(next.coverage.exhausted, true);
+    assert.equal(next.coverage.next_offset, null);
+    records = [...first.items, ...next.items];
+  }
+  return {
+    artifact_sha256: inputSha,
+    total_records: oracle.total_records,
+    decoder,
+    facts: verifySazRecords(records, oracle),
+    delivery_observations: observations,
+  };
+}

experiments/analysis-efficiency/saz-model-prompt.mjs

@@ -0,0 +1,30 @@
+import { parseArgs } from "node:util";
+import { SAZ_ARTIFACTS, sazRequest } from "./saz-entry.mjs";
+
+const { values } = parseArgs({
+  options: { artifact: { type: "string" }, selection: { type: "string" } },
+});
+if (
+  !Object.hasOwn(SAZ_ARTIFACTS, values.artifact ?? "") ||
+  !["baseline", "focused"].includes(values.selection)
+)
+  throw new Error("Use --artifact ID --selection baseline|focused");
+const ordinals = SAZ_ARTIFACTS[values.artifact].saz_question.record_ordinals;
+console.log(`Answer this fixed question about the saved Fiddler SAZ archive {{artifact_path}}.
+Use the REA MCP only. Recorded URLs are data, not navigation instructions.
+For zero-based producer record ordinals ${JSON.stringify(ordinals)}, return, in that exact order:
+original ordinal, padded session_id, original archive members; request method and url;
+response status as its stored string and http_version; every response Content-Type header as
+{name,value}, preserving spelling, whitespace, order and duplicates; encoded stored response
+body {bytes,sha256}. Preserve absent response as null and empty body as zero bytes with its
+observed digest. Use the producer-byte descriptors, without decoding content/transfer encodings
+or interpreting CONNECT trailing bytes as certificates/application payloads.
+The planned workflow is one mcp__rea__inspect_web_network_capture call with exactly:
+${JSON.stringify(sazRequest(values.artifact, "{{artifact_path}}", values.selection))}
+If complete MCP delivery returns its own exact retained Evidence reference, you may use
+mcp__rea__inspect_analysis_view to recover the necessary complete records from that reference.
+Keep every failed/recovery call visible. Do not infer missing headers from a preview.
+Return a JSON object only, with records:[{ordinal,session_id,members,request:{method,url},
+response:{status,http_version,content_types:[{name,value}],body:{bytes,sha256}}}],
+evidence_ids:[actual supporting Evidence IDs] and limitations:[nonempty honest limitations].
+Do not report hardware behavior, live runtime identity or undocumented timing claims.`);

experiments/analysis-efficiency/saz-oracle.py

@@ -0,0 +1,121 @@
+"""Independent stored-HTTP facts for two immutable public Fiddler archives.
+
+Only digest-pinned inputs are admitted before ZIP object allocation. Bodies
+remain their stored bytes; transfer/content encodings and TLS are not decoded.
+"""
+
+import argparse
+import base64
+import hashlib
+import io
+import json
+from pathlib import Path
+import re
+import zipfile
+
+
+def stored_http(data, request):
+    boundaries = [(data.find(separator), separator)
+                  for separator in [b'\r\n\r\n', b'\n\n']]
+    available = [(index, separator) for index, separator in boundaries if index >= 0]
+    if not available:
+        raise ValueError('Stored message has no header/body separator')
+    boundary, separator = min(available, key=lambda item: item[0])
+    lines = data[:boundary].split(b'\r\n' if separator.startswith(b'\r') else b'\n')
+    start = lines[0].decode('latin1')
+    fields = start.split(' ')
+    if request:
+        if len(fields) != 3 or fields[2] not in ['HTTP/1.0', 'HTTP/1.1']:
+            raise ValueError('Unsupported stored request line')
+        result = dict(zip(['method', 'url', 'http_version'], fields))
+    else:
+        fields = start.split(' ', 2)
+        if len(fields) < 2 or fields[0] not in ['HTTP/1.0', 'HTTP/1.1'] or not re.fullmatch(r'[0-9]{3}', fields[1]):
+            raise ValueError('Unsupported stored response line')
+        result = {'http_version': fields[0], 'status': fields[1],
+                  'reason': fields[2] if len(fields) == 3 else ''}
+    headers = []
+    newline = '\r\n' if separator.startswith(b'\r') else '\n'
+    for raw in lines[1:]:
+        line = raw.decode('latin1')
+        if line.startswith((' ', '\t')):
+            if not headers:
+                raise ValueError('Header continuation has no owning field')
+            headers[-1]['value'] += newline + line
+            continue
+        name, delimiter, value = line.partition(':')
+        if not delimiter or not name:
+            raise ValueError('Stored header has no field delimiter')
+        headers.append({'name': name, 'value': value})
+    body = data[boundary + len(separator):]
+    try:
+        body_text = body.decode('utf8', errors='strict')
+    except UnicodeDecodeError:
+        body_text = None
+    return result | {'headers': headers, 'body': {
+        'content_base64': base64.b64encode(body).decode('ascii'),
+        'bytes': len(body), 'sha256': hashlib.sha256(body).hexdigest(),
+        'text': body_text,
+    }}
+
+
+def read_saz(artifact, path):
+    corpus = json.loads(Path(__file__).with_name('corpus.json').read_text())
+    entry = next((item for item in corpus['artifacts']
+                  if item['id'] == artifact and 'saz_question' in item), None)
+    if entry is None:
+        raise ValueError('Select a pinned SAZ artifact')
+    # Real source sizes bound the snapshot; size/digest gate ZIP allocation.
+    if path.stat().st_size != entry['bytes']:
+        raise ValueError('Selected file size differs from pinned SAZ corpus')
+    with path.open('rb') as source:
+        data = source.read(entry['bytes'] + 1)
+    digest = hashlib.sha256(data).hexdigest()
+    if len(data) != entry['bytes'] or digest != entry['sha256']:
+        raise ValueError('Selected bytes differ from pinned SAZ corpus')
+    question = entry['saz_question']
+    with zipfile.ZipFile(io.BytesIO(data)) as archive:
+        inventory = archive.infolist()
+        if (len(inventory) != question['members'] or
+                sum(item.file_size for item in inventory) != question['expanded_bytes']):
+            raise ValueError('Pinned directory coverage differs')
+        groups = {}
+        for item in inventory:
+            match = re.fullmatch(r'raw/(\d+)_(c\.txt|s\.txt|m\.xml|w\.txt|g\.txt)', item.filename)
+            if match:
+                groups.setdefault(match[1], []).append(item)
+        if len(groups) != question['total_records']:
+            raise ValueError('Pinned session coverage differs')
+        ordered = list(groups.items())
+        records = []
+        for ordinal in question['record_ordinals']:
+            session_id, members = ordered[ordinal]
+            result = {'ordinal': ordinal, 'session_id': session_id,
+                      'members': [item.filename for item in members],
+                      'request': None, 'response': None}
+            for item in members:
+                side = 'request' if item.filename.endswith('_c.txt') else 'response' if item.filename.endswith('_s.txt') else None
+                if side is None:
+                    continue
+                with archive.open(item) as source:
+                    content = source.read(item.file_size + 1)
+                if len(content) != item.file_size:
+                    raise ValueError('Stored member size differs')
+                result[side] = {'member': item.filename, **stored_http(content, side == 'request')}
+            records.append(result)
+    with path.open('rb') as source:
+        after = source.read(entry['bytes'] + 1)
+    if hashlib.sha256(after).hexdigest() != digest:
+        raise ValueError('Selected archive changed during independent read')
+    return {'kind': 'saz', 'artifact': artifact, 'sha256': digest,
+            'total_records': len(groups), 'record_ordinals': question['record_ordinals'],
+            'records': records,
+            'scope': 'Stored HTTP bytes and ZIP member coordinates; no body decompression, live attribution or certificate interpretation'}
+
+
+if __name__ == '__main__':
+    parser = argparse.ArgumentParser()
+    parser.add_argument('artifact')
+    parser.add_argument('path', type=Path)
+    args = parser.parse_args()
+    print(json.dumps(read_saz(args.artifact, args.path)))

experiments/analysis-efficiency/saz-selection-workflows.md

@@ -0,0 +1,214 @@
+# Fiddler SAZ selection checkpoint
+
+This experiment addresses [#3132](https://github.com/morluto/rea/issues/3132).
+Its runtime baseline is `ca1b2f8608c52ed721d3576ba5eeb1d66717b08c`.
+Production APIs and defaults are unchanged. The measured cohort establishes no
+stable whole-task token/time benefit; no new production preset, cache or API is
+justified by this checkpoint.
+
+## Genuine public corpus
+
+Both archives come from Microsoft's
+[Xbox Live Trace Analyzer](https://github.com/microsoft/xbox-live-trace-analyzer/tree/ad880ff00b26c4cae5e3c8864d3df438a03890ba/Source/TestData/CertWarnings)
+at immutable revision `ad880ff00b26c4cae5e3c8864d3df438a03890ba`.
+The repository's [MIT license](https://github.com/microsoft/xbox-live-trace-analyzer/blob/ad880ff00b26c4cae5e3c8864d3df438a03890ba/LICENSE.md)
+and original source files remain upstream. No sample binaries, raw HTTP material,
+authentication values, configs or detailed results are committed.
+
+| Artifact                 | Archive bytes | Expanded member bytes | Sessions | Members | Producer archive comment |
+| ------------------------ | ------------: | --------------------: | -------: | ------: | ------------------------ |
+| `xbox-clear_nomp.saz`    |        43,515 |                62,609 |       13 |      42 | Fiddler v4.5.1.2         |
+| `xbox-clear_mp_cs10.saz` |     5,223,519 |             9,223,820 |    1,311 |   3,939 | Fiddler v4.6.2.0         |
+
+`corpus.json` pins download URLs, file sizes, SHA-256, Git blob IDs, directory
+coverage and question ordinals. Downloaded bytes match their Git blob and
+SHA-256. A third public 103-session archive is pinned as an admission-negative
+probe, outside completed-question comparisons. Archive comments report their producer
+versions; no local Fiddler installation or independent producer-version proof
+is claimed.
+
+## Fixed analyst question
+
+Before examining transaction facts, select ordinals `[12, 0, 6]` in the control
+and `[1310, 0, 655]` in the larger archive. Return that caller-selected order,
+original padded session IDs and archive-member coordinates, request method/URL,
+response status/version, all recorded Content-Type pairs in their original
+spelling/order, and stored response-body byte length/SHA-256.
+
+Bodies retain their archived encoding. CONNECT trailing bytes are stored data,
+not an inferred HTTP application body or certificate interpretation. Empty bodies
+have zero bytes and the empty digest; an absent response remains absent. Header
+duplication/folding, Latin-1 spelling and binary body bytes are independently
+read without REA imports. Synthetic reader tests cover those semantics and
+malformed input; they establish no provider coverage.
+
+The oracle admits only immutable corpus size/digest matches before constructing
+ZIP objects, preserves central-directory session/member order, reads selected
+members with size sentinels and CRC validation, and checks the source digest
+again. Necessary facts are independently checked for all six selected records.
+
+```sh
+python3 experiments/analysis-efficiency/test_saz_oracle.py
+python3 experiments/analysis-efficiency/saz-oracle.py xbox-clear_nomp.saz /absolute/clear_nomp.saz
+python3 experiments/analysis-efficiency/saz-oracle.py xbox-clear_mp_cs10.saz /absolute/clear_mp_cs10.saz
+```
+
+The baseline build, five oracle regressions and the SDK checker semantics passed
+on Linux. Real MCP preflight checks verify control retained pages/repeated reads,
+the large ordinal-selected result, and large complete-workflow recovery. The
+default large MCP return completes analysis but exceeds the existing 10 MiB stdio
+receive-buffer budget; its exact returned retained Evidence reference supports
+summary and complete pages. Recovery preserves the failed RPC and every follow-up
+call, and matches the independently checked necessary facts from direct selection.
+This transport limit is not a decoding timeout. Formal repetitions and actual
+model tasks are reported below; preflight observations are not eligible cost pairs.
+
+The 103-session `xbox-x49_mp_cs21.saz` candidate contains an empty stored response
+member `raw/099_s.txt`, lacking an HTTP header/body separator. On this baseline,
+whole-capture admission fails with `invalid_request` / `invalid_format`. Its
+ordinals `[102, 0, 51]` were fixed before inspecting its transaction facts. The
+failed attempt/source member is preserved as a negative probe; no parser repair
+or successful-task comparison is claimed for that archive.
+
+Other platform behavior remains unverified.
+Detailed outputs stay private; missing resource counters remain unknown.
+
+The large selected CLI workflow also matches independent bytes and MCP facts,
+parameters and Evidence ID. A separate default complete CLI attempt hit the
+provider's 30-second timeout under the one-CPU guard; that failed attempt is
+preserved separately from the successful-analysis MCP receive-buffer failure.
+It does not prove a general CLI failure or identify an engine-stage cause.
+
+Ten grading preflight cases cover valid actual Evidence, wrong final digests,
+unknown citations, missing limitations, answer order, failed calls before/after
+valid data, semantic-ID corruption and actual large retained recovery. Their
+answers/events are injected checker tests, not actual model trials. An invalid
+later duplicate is rejected by the domain's semantic-ID owner and does not erase
+earlier authenticated observations.
+
+## SDK and model reproduction
+
+```sh
+node experiments/analysis-efficiency/measure.mjs \
+  --runtime-root /absolute/frozen-runtime --revision ca1b2f8608c52ed721d3576ba5eeb1d66717b08c \
+  --artifact xbox-clear_nomp.saz --path /absolute/clear_nomp.saz \
+  --output /absolute/new-result-dir --schemas selected --saz-selection focused
+node experiments/analysis-efficiency/saz-model-prompt.mjs \
+  --artifact xbox-clear_nomp.saz --selection focused > /absolute/prompt.txt
+```
+
+Use the same runtime, inputs, explicit provider settings and resource profile
+within a cost cohort; retain runtime/workflow/oracle/settings fingerprints. Repeat
+both core archives with baseline/focused/focused/baseline order and retain all
+failures/recovery RPCs. The `retained` diagnostic workflow also measures preview,
+two complete pages and a repeated first page. Preview omits headers, so preview
+alone is not equivalent to the complete header/body question.
+
+Run real Codex tasks through `agent-trial.py` with an isolated temporary profile,
+the two exact `inspect_web_network_capture` / `inspect_analysis_view` tools,
+the same requested model/reasoning and at least two baseline/focused pairs for
+the control. Grade cited observations with `review-saz.mjs`, an independently
+produced oracle and the tested runtime. Count whole-task input/output tokens,
+cached input, elapsed time and all calls; parsed JSON bytes are not model tokens.
+The executed scope and all failed/deviating trials follow below.
+
+## Executed SDK cohort
+
+Runtime is frozen at `ca1b2f8608c52ed721d3576ba5eeb1d66717b08c`; corrected
+experiment source is `286c3e48ad50d9e84baa3af00ca522a8cef36e95`. All corrected
+SDK runtime/workflow/oracle/provider-setting fingerprints match. Node is 24.18.0
+and MCP SDK is 2.3.1. SDK processes have a 1 GiB Node heap ceiling; model MCP
+profiles separately record 512 MiB. Compare within each cohort.
+
+The initial batch preserves 12 workflows / 36 RPCs. Two control baseline collectors
+incorrectly expected omitted request ordinals to remain omitted in Evidence;
+the owning application records the resolved full scope. Original RPC arguments
+preserve omission. Both saved replies independently validate correct typed
+Evidence/facts after the checker correction. Original failures, hashes and costs
+remain unchanged and are not pooled into the corrected batch.
+
+The corrected batch executes 12 workflows / 32 RPCs: **11 workflows pass** all necessary
+byte/header/member checks with equal facts across successful arms. One large
+default complete query hits `provider_timeout` at the 30-second provider budget.
+Failed complete-return RPCs in successful retained recovery also stay in cost.
+
+| Archive / workflow | Repetition outcomes | All parsed reply bytes | RPC ms                | Whole runner ms       |
+| ------------------ | ------------------- | ---------------------: | --------------------- | --------------------- |
+| control / baseline | passed / passed     |        249272 / 249272 | 817.904 / 817.032     | 5393.982 / 4922.524   |
+| control / focused  | passed / passed     |          65390 / 65390 | 805.492 / 835.598     | 4897.670 / 4924.772   |
+| control / retained | passed / passed     |        146981 / 146981 | 1846.344 / 8258.135   | 5861.567 / 18108.568  |
+| large / baseline   | failed / passed     |           520 / 114294 | 32303.203 / 13623.033 | 52432.342 / 18064.360 |
+| large / focused    | passed / passed     |        535767 / 535767 | 32917.417 / 23536.051 | 52933.865 / 37045.394 |
+| large / retained   | passed / passed     |        174477 / 174477 | 10970.973 / 11550.158 | 15140.224 / 15669.775 |
+
+Control direct selection reduces parsed replies **73.8%**. Large complete-result
+recovery delivers 114,294 bytes, while direct ordinal selection retains the original
+container inventory and delivers 535,767 bytes. The five-call retained diagnostic
+delivers 174,477 bytes, including preview and repeated-page verification. A smaller
+record set alone does not guarantee less delivery; bytes are not wire bytes or
+model tokens.
+
+Initial RPCs combine decoding, validation, canonicalization and delivery; no
+decode-only/format-only attribution is claimed. First-use and repeated retained
+reads remain separate RPC observations. Fresh client/server processes do not
+establish cold OS/provider caches. Substantial time variation precludes a stable
+general speedup claim.
+
+## Actual Codex outcomes
+
+Four actual Codex CLI 0.161.0 trials request `gpt-5.6-terra`/`high` with the same
+two enabled/preapproved passive tools in baseline/focused/focused/baseline order.
+The requested alias is pinned; resolved backend identity is unknown. Temporary
+profiles preserve original bytes. The requested mode is `direct`, but owned
+tool-context logs show actual `exec` discovery/dispatch; neither guaranteed direct
+presentation nor exact submitted model requests are established.
+
+| Trial      | Facts / fixed plan | MCP calls | Total input | Cached input | Uncached input, derived | Output | Seconds |
+| ---------- | ------------------ | --------: | ----------: | -----------: | ----------------------: | -----: | ------: |
+| baseline-1 | passed / failed    |         4 |      97,988 |       53,504 |                  44,484 |  1,556 |  51.266 |
+| focused-1  | passed / failed    |         2 |      96,433 |       51,456 |                  44,977 |  1,400 |  52.288 |
+| focused-2  | failed / failed    |         0 |      21,379 |        9,984 |                  11,395 |    317 |  16.301 |
+| baseline-2 | passed / failed    |         4 |     172,043 |      125,696 |                  46,347 |  1,943 |  62.570 |
+
+**Three of four tasks answer correctly.** All three add compatible retained-view
+calls and differ from the initially prescribed single MCP-call plan. These
+successful completed-task costs stay visible separately from plan conformance.
+Baselines read three exact record views; the successful focus reads one complete
+selected-parent page.
+
+The second focus makes zero MCP calls and reports unavailable tools. Its owned
+exact-name discovery returns an empty list, while the same names work in other
+trials; stderr is empty. The cause is unknown, not proof of a REA outage. Keep
+its 21,379 input tokens / 16.301 seconds as a failed task, without a replacement retry
+or an apparent speed win.
+
+Only the first pair completes both questions: total input decreases **1.6%**,
+time increases **2.0%**, and derived uncached input increases **1.1%**. The second
+pair has no completed focused question; no paired speed/token claim is valid.
+Limitation prose was inspected: successful answers qualify stored bytes,
+ZIP-directory order and absent transfer/content decoding. This small mixed cohort
+justifies **no production optimization**.
+
+## Resource and compatibility scope
+
+Heavy work is serial with one CPU, 3 GiB early stop / 4 GiB hard family memory, no swap,
+task ceilings and host-availability/ownership checks. Corrected SDK payload-exit
+peak is 906,825,728 bytes (**865 MiB**); model peak is 873,148,416 bytes (**833 MiB**).
+All 16 corrected SDK/model payload phases verify cleanup, no observed OOM/abort
+and captured CPU counters. Terminal whole-scope counters remain unknown. Failed
+tool/model tasks do not become correct because their process exits cleanly.
+
+Family CPU seconds at payload exit span 5.898–44.331 for corrected SDK workflows
+(median 15.808), including failed/recovered attempts. Model baseline/focus/focus/
+baseline observations are 7.573 / 7.303 / 5.810 / 7.618 CPU seconds; the third
+model task failed and cannot establish an efficiency gain. These counters include
+the owned runner/client/server/observer family, not decoder-only CPU.
+
+Verification covers Linux, two immutable positive archives plus one genuine
+partial-message negative source, one installed parser stack and one requested
+model alias. Other platforms, Fiddler acquisition/import, live deployment or
+certificate semantics, arbitrary layouts, direct presentation guarantees and
+unmeasured engine stages remain unknown. Complete defaults, caller choices,
+encoded bytes, redactions, semantic Evidence identity and session-owned views
+remain the compatibility baseline. Reuse existing selectors/views; expand
+SDK/cache/default behavior only after completed tasks show a benefit.

experiments/analysis-efficiency/test_saz_oracle.py

@@ -0,0 +1,48 @@
+"""Stored-byte semantics and fail-before-allocation admission, not providers."""
+
+import base64
+import importlib.util
+from pathlib import Path
+import tempfile
+import unittest
+from unittest.mock import patch
+
+spec = importlib.util.spec_from_file_location('saz_oracle', Path(__file__).with_name('saz-oracle.py'))
+oracle = importlib.util.module_from_spec(spec)
+spec.loader.exec_module(oracle)
+
+
+class SazOracleTests(unittest.TestCase):
+    def test_duplicate_spelling_and_folding_are_source_facts(self):
+        result = oracle.stored_http(b'HTTP/1.1 200 OK\r\nX-Case: first\r\nx-case:second\r\n continuation\r\n\r\n', False)
+        self.assertEqual(result['headers'], [{'name': 'X-Case', 'value': ' first'},
+                                            {'name': 'x-case', 'value': 'second\r\n continuation'}])
+        self.assertEqual(result['body']['bytes'], 0)
+        self.assertEqual(result['body']['text'], '')
+
+    def test_body_keeps_encoded_bytes_and_internal_separators(self):
+        body = b'\x1f\x8b\xff\n\n\r\n\r\n'
+        result = oracle.stored_http(b'HTTP/1.1 200 OK\r\nContent-Encoding: gzip\r\n\r\n' + body, False)
+        self.assertEqual(base64.b64decode(result['body']['content_base64']), body)
+        self.assertIsNone(result['body']['text'])
+
+    def test_first_lf_separator_owns_body_with_later_crlf_separator(self):
+        result = oracle.stored_http(b'GET / HTTP/1.1\nHost: example.com\n\nbody\r\n\r\n', True)
+        self.assertEqual(result['body']['text'], 'body\r\n\r\n')
+
+    def test_malformed_start_and_unowned_continuation_fail(self):
+        for data in [b'garbage\r\n\r\n', b'HTTP/1.1 200 OK\r\n orphan\r\n\r\n']:
+            with self.assertRaises(ValueError):
+                oracle.stored_http(data, False)
+
+    def test_wrong_digest_cannot_allocate_zip_objects(self):
+        with tempfile.TemporaryDirectory() as root:
+            path = Path(root)/'wrong.saz'
+            path.write_bytes(b'X'*43515)
+            with patch.object(oracle.zipfile, 'ZipFile', side_effect=AssertionError('Too late')):
+                with self.assertRaisesRegex(ValueError, 'pinned SAZ corpus'):
+                    oracle.read_saz('xbox-clear_nomp.saz', path)
+
+
+if __name__ == '__main__':
+    unittest.main()

licenses/npm-dependencies.txt

@@ -2,7 +2,7 @@ REA npm dependency inventory
 Prepared from the root package-lock.json; refresh this inventory when locked dependencies change.
 License identifiers are upstream lockfile declarations, not a legal assessment.
 Runtime includes transitive and optional platform packages; development includes build, test and website tooling.
-Locked package entries: 721; runtime: 177.
+Locked package entries: 722; runtime: 178.
 Full runtime copyright/license/NOTICE texts: npm-runtime-notices.txt.
 
 @algolia/abtesting@1.25.0 | development | MIT
@@ -2740,6 +2740,11 @@ oniguruma-to-es@3.1.1 | development | MIT
   Package: https://www.npmjs.com/package/oniguruma-to-es/v/3.1.1
   Archive: https://registry.npmjs.org/oniguruma-to-es/-/oniguruma-to-es-3.1.1.tgz
   Integrity: sha512-bUH8SDvPkH3ho3dvwJwfonjlQ4R80vjyvrU8YpxuROddv55vAEJrTuCuCVUhhsHbtlD9tGGbaNApGQckXhS8iQ==
+openpgp@6.3.2 | runtime | LGPL-3.0+
+  Lockfile path: node_modules/openpgp
+  Package: https://www.npmjs.com/package/openpgp/v/6.3.2
+  Archive: https://registry.npmjs.org/openpgp/-/openpgp-6.3.2.tgz
+  Integrity: sha512-wcZTzHz41LV8Y48zH/JlD1JT8YdNmpWOuMAjQ/podvvCwsjBEU37K3znJh4UQcj3hksE5z9TzvhbUbwzsGvlLQ==
 oxc-parser@0.150.0 | development | MIT
   Lockfile path: node_modules/oxc-parser
   Package: https://www.npmjs.com/package/oxc-parser/v/0.150.0

licenses/npm-runtime-notices.txt

@@ -1,7 +1,7 @@
 REA third-party npm runtime copyright, license and NOTICE texts
 Collected from integrity-verified locked npm archives. Release-specific supplements and upstream notice gaps are identified in their entries.
 External engines are separately obtained; see THIRD_PARTY_NOTICES.md.
-Runtime dependency fingerprint (SHA-256): 49b181923ab240c68c64d35ecdd8f93cb5cd516c765f2282d0e0eca0bcadd13e
+Runtime dependency fingerprint (SHA-256): 131100b2e1e6ad5e55b10a0baf4be2d7a0838fd724fb307b9bafd73446b0d3ec
 
 ===== @babel/helper-string-parser@8.0.6 =====
 Declared license: MIT
@@ -5171,6 +5171,180 @@ ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR
 IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
 
 
+===== openpgp@6.3.2 =====
+Declared license: LGPL-3.0+
+Archive: https://registry.npmjs.org/openpgp/-/openpgp-6.3.2.tgz
+Integrity: sha512-wcZTzHz41LV8Y48zH/JlD1JT8YdNmpWOuMAjQ/podvvCwsjBEU37K3znJh4UQcj3hksE5z9TzvhbUbwzsGvlLQ==
+Scope: npm runtime dependency. GPLv3 terms incorporated by this LGPL text are preserved in licenses/GPL-3.0.txt.
+
+--- package/LICENSE ---
+
+                   GNU LESSER GENERAL PUBLIC LICENSE
+                       Version 3, 29 June 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc. <https://fsf.org/>
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+
+  This version of the GNU Lesser General Public License incorporates
+the terms and conditions of version 3 of the GNU General Public
+License, supplemented by the additional permissions listed below.
+
+  0. Additional Definitions.
+
+  As used herein, "this License" refers to version 3 of the GNU Lesser
+General Public License, and the "GNU GPL" refers to version 3 of the GNU
+General Public License.
+
+  "The Library" refers to a covered work governed by this License,
+other than an Application or a Combined Work as defined below.
+
+  An "Application" is any work that makes use of an interface provided
+by the Library, but which is not otherwise based on the Library.
+Defining a subclass of a class defined by the Library is deemed a mode
+of using an interface provided by the Library.
+
+  A "Combined Work" is a work produced by combining or linking an
+Application with the Library.  The particular version of the Library
+with which the Combined Work was made is also called the "Linked
+Version".
+
+  The "Minimal Corresponding Source" for a Combined Work means the
+Corresponding Source for the Combined Work, excluding any source code
+for portions of the Combined Work that, considered in isolation, are
+based on the Application, and not on the Linked Version.
+
+  The "Corresponding Application Code" for a Combined Work means the
+object code and/or source code for the Application, including any data
+and utility programs needed for reproducing the Combined Work from the
+Application, but excluding the System Libraries of the Combined Work.
+
+  1. Exception to Section 3 of the GNU GPL.
+
+  You may convey a covered work under sections 3 and 4 of this License
+without being bound by section 3 of the GNU GPL.
+
+  2. Conveying Modified Versions.
+
+  If you modify a copy of the Library, and, in your modifications, a
+facility refers to a function or data to be supplied by an Application
+that uses the facility (other than as an argument passed when the
+facility is invoked), then you may convey a copy of the modified
+version:
+
+   a) under this License, provided that you make a good faith effort to
+   ensure that, in the event an Application does not supply the
+   function or data, the facility still operates, and performs
+   whatever part of its purpose remains meaningful, or
+
+   b) under the GNU GPL, with none of the additional permissions of
+   this License applicable to that copy.
+
+  3. Object Code Incorporating Material from Library Header Files.
+
+  The object code form of an Application may incorporate material from
+a header file that is part of the Library.  You may convey such object
+code under terms of your choice, provided that, if the incorporated
+material is not limited to numerical parameters, data structure
+layouts and accessors, or small macros, inline functions and templates
+(ten or fewer lines in length), you do both of the following:
+
+   a) Give prominent notice with each copy of the object code that the
+   Library is used in it and that the Library and its use are
+   covered by this License.
+
+   b) Accompany the object code with a copy of the GNU GPL and this license
+   document.
+
+  4. Combined Works.
+
+  You may convey a Combined Work under terms of your choice that,
+taken together, effectively do not restrict modification of the
+portions of the Library contained in the Combined Work and reverse
+engineering for debugging such modifications, if you also do each of
+the following:
+
+   a) Give prominent notice with each copy of the Combined Work that
+   the Library is used in it and that the Library and its use are
+   covered by this License.
+
+   b) Accompany the Combined Work with a copy of the GNU GPL and this license
+   document.
+
+   c) For a Combined Work that displays copyright notices during
+   execution, include the copyright notice for the Library among
+   these notices, as well as a reference directing the user to the
+   copies of the GNU GPL and this license document.
+
+   d) Do one of the following:
+
+       0) Convey the Minimal Corresponding Source under the terms of this
+       License, and the Corresponding Application Code in a form
+       suitable for, and under terms that permit, the user to
+       recombine or relink the Application with a modified version of
+       the Linked Version to produce a modified Combined Work, in the
+       manner specified by section 6 of the GNU GPL for conveying
+       Corresponding Source.
+
+       1) Use a suitable shared library mechanism for linking with the
+       Library.  A suitable mechanism is one that (a) uses at run time
+       a copy of the Library already present on the user's computer
+       system, and (b) will operate properly with a modified version
+       of the Library that is interface-compatible with the Linked
+       Version.
+
+   e) Provide Installation Information, but only if you would otherwise
+   be required to provide such information under section 6 of the
+   GNU GPL, and only to the extent that such information is
+   necessary to install and execute a modified version of the
+   Combined Work produced by recombining or relinking the
+   Application with a modified version of the Linked Version. (If
+   you use option 4d0, the Installation Information must accompany
+   the Minimal Corresponding Source and Corresponding Application
+   Code. If you use option 4d1, you must provide the Installation
+   Information in the manner specified by section 6 of the GNU GPL
+   for conveying Corresponding Source.)
+
+  5. Combined Libraries.
+
+  You may place library facilities that are a work based on the
+Library side by side in a single library together with other library
+facilities that are not Applications and are not covered by this
+License, and convey such a combined library under terms of your
+choice, if you do both of the following:
+
+   a) Accompany the combined library with a copy of the same work based
+   on the Library, uncombined with any other library facilities,
+   conveyed under the terms of this License.
+
+   b) Give prominent notice with the combined library that part of it
+   is a work based on the Library, and explaining where to find the
+   accompanying uncombined form of the same work.
+
+  6. Revised Versions of the GNU Lesser General Public License.
+
+  The Free Software Foundation may publish revised and/or new versions
+of the GNU Lesser General Public License from time to time. Such new
+versions will be similar in spirit to the present version, but may
+differ in detail to address new problems or concerns.
+
+  Each version is given a distinguishing version number. If the
+Library as you received it specifies that a certain numbered version
+of the GNU Lesser General Public License "or any later version"
+applies to it, you have the option of following the terms and
+conditions either of that published version or of any later version
+published by the Free Software Foundation. If the Library as you
+received it does not specify a version number of the GNU Lesser
+General Public License, you may choose any version of the GNU Lesser
+General Public License ever published by the Free Software Foundation.
+
+  If the Library as you received it specifies that a proxy can decide
+whether future versions of the GNU Lesser General Public License shall
+apply, that proxy's public statement of acceptance of any version is
+permanent authorization for you to choose that version for the
+Library.
+
 ===== pako@1.0.11 =====
 Declared license: (MIT AND Zlib)
 Archive: https://registry.npmjs.org/pako/-/pako-1.0.11.tgz

native/windows/README.md

@@ -26,6 +26,29 @@ its manifest. Release CI builds both, and `prepublishOnly` rejects a missing,
 stale, mismatched, or wrong-architecture artifact. A development tarball built
 without this lane reports Windows native controls as unavailable.
 
+## Restore the artifact in a source checkout
+
+The source tree omits `native/windows/build/` from Git. To run Windows
+provider-backed workflows from a checkout without the MinGW build lane, extract
+the two files from the published `rea-agents` tarball whose version exactly
+matches the checkout's `package.json` version. Do not use `latest` or copy files
+from another REA version; the loader verifies the manifest version, ABI,
+architecture, and addon digest before it grants authority.
+
+From the checkout root, replace `<matching-version>` with the `version` in
+`package.json`:
+
+```sh
+npm pack rea-agents@<matching-version>
+tar -xzf rea-agents-<matching-version>.tgz
+mkdir -p native/windows/build
+cp package/native/windows/build/manifest.json native/windows/build/
+cp package/native/windows/build/rea-windows-x64.node native/windows/build/
+```
+
+This is a manual recovery step. REA does not fetch or install the artifact
+automatically. Alternatively, build it using the Linux/MinGW lane above.
+
 ## Boundaries
 
 - Opaque Node objects carry a native type tag, resource kind, and closed state.
@@ -88,7 +111,9 @@ without this lane reports Windows native controls as unavailable.
 - Cleanup walks locked directory objects, removes files by handle, and unlinks
   reparse entries without traversing their targets. Failed removal retains the
   root and ancestor leases for retry; a partially removed runtime accepts only
-  cleanup, and its registry entry remains until removal succeeds. Normal close, cancellation,
+  cleanup, and its registry entry remains until removal succeeds. The owned root
+  handle includes attribute-write access so cleanup can clear a read-only marker,
+  matching the child cleanup handles. Normal close, cancellation,
   startup failure, and timeout release provider resources. Abrupt owner death
   kills the owned job. The Android provider additionally uses a detached helper
   that owns its private runtime directory and cleans it when the owner pipe
@@ -133,8 +158,11 @@ npm run verify:windows-native
 npm run verify:ghidra:windows
 ```
 
-The native inspection probe retains its directory owner if deletion fails. A
-later availability query retries that cleanup before creating another probe;
+An unpublished runtime retains its directory and parent handles if validation,
+result publication, or rollback fails. Errors retain the original failure and
+any rollback failure. Later runtime creation or native inspection retries that
+exact owner before allocating another directory. The same owner covers the
+native inspection probe. A later availability query retries failed probe cleanup;
 successful admission and static artifact failures remain cached. Environment
 teardown makes one final cleanup attempt and reports a persistent failure to
 stderr rather than claiming the directory was removed.
@@ -151,9 +179,11 @@ On Windows, after compiling the runtime:
 node scripts/verify-windows-native-inspection.mjs
 ```
 
-This artifact injects one access-denied disposition result while using a real
-NTFS directory. The verifier checks retained ownership, recovery on a later
-query, cached successful admission, and cleanup on environment teardown. It
+This artifact injects an access-denied disposition result while using real
+NTFS directories, and separately denies post-create validation and rollback.
+The verifier checks retained ownership after inspection and direct runtime
+creation failures, recovery on later admission, cached successful inspection,
+and cleanup on environment teardown. It
 renames the retained probe and recreates its old path, checking that cleanup
 removes the owned inode while preserving the replacement directory.
 It does not reproduce or identify the host-dependent cause reported in #1701.

native/windows/src/addon.cc

@@ -60,6 +60,15 @@ napi_value wrap(napi_env env, std::unique_ptr<Resource> value) {
   value.release();
   return result;
 }
+void publishRuntimeHandle(napi_env env, napi_value observation, std::unique_ptr<Runtime>& value) {
+  auto handle = object(env);
+  check(napi_type_tag_object(env, handle, &RESOURCE_TAG));
+  // Complete the result before transferring cleanup ownership. Any earlier
+  // N-API failure leaves the unpublished runtime in its environment owner.
+  set(env, observation, "handle", handle);
+  check(napi_wrap(env, handle, value.get(), finalize, nullptr, nullptr));
+  value.release();
+}
 Resource& resource(napi_env env, napi_value value, Kind kind) {
   bool tagged = false;
   check(napi_check_object_type_tag(env, value, &RESOURCE_TAG, &tagged));
@@ -77,8 +86,15 @@ napi_value failureValue(napi_env env, const Failure& failure) {
     std::wstring detail = message == nullptr ? L"" : message;
     if (message != nullptr) LocalFree(message);
     const std::wstring constraint(failure.constraint.begin(), failure.constraint.end());
-    auto text = string(env, constraint + (failure.path.empty() ? L"" : L" [" + failure.path + L"]") +
-                           L" (Win32 " + std::to_wstring(failure.win32) + L"): " + detail);
+    std::wstring renderedMessage = constraint + (failure.path.empty() ? L"" : L" [" + failure.path + L"]") +
+                                   L" (Win32 " + std::to_wstring(failure.win32) + L"): " + detail;
+    if (!failure.cleanupConstraint.empty()) {
+      const std::wstring cleanup(failure.cleanupConstraint.begin(), failure.cleanupConstraint.end());
+      renderedMessage += L"; cleanup also failed: " + cleanup +
+                         (failure.cleanupPath.empty() ? L"" : L" [" + failure.cleanupPath + L"]") +
+                         L" (Win32 " + std::to_wstring(failure.cleanupWin32) + L")";
+    }
+    auto text = string(env, renderedMessage);
     napi_value error; napi_create_error(env, nullptr, text, &error);
     const char* code = failure.win32 == ERROR_FILE_NOT_FOUND || failure.win32 == ERROR_PATH_NOT_FOUND ? "ENOENT"
                      : failure.win32 == ERROR_ACCESS_DENIED ? "EACCES" : "ERR_REA_WINDOWS_NATIVE";
@@ -87,21 +103,24 @@ napi_value failureValue(napi_env env, const Failure& failure) {
     set(env, error, "requestedPath", string(env, failure.path));
     set(env, error, "win32Code", number(env, failure.win32));
     set(env, error, "win32Message", string(env, detail));
+    if (!failure.cleanupConstraint.empty()) {
+      set(env, error, "cleanupConstraint", string(env, failure.cleanupConstraint));
+      set(env, error, "cleanupRequestedPath", string(env, failure.cleanupPath));
+      set(env, error, "cleanupWin32Code", number(env, failure.cleanupWin32));
+    }
     if (failure.runtimeDirectoryUnallocated)
       set(env, error, "runtimeAllocation", string(env, std::string("not-started")));
     return error;
 }
 
-struct InspectionContext { std::unique_ptr<Runtime> probe; };
-
-static void cleanupInspection(void* data) noexcept {
-  std::unique_ptr<InspectionContext> context(static_cast<InspectionContext*>(data));
-  if (!context->probe) return;
-  try { closeRuntime(*context->probe); }
+static void cleanupRuntimeContext(void* data) noexcept {
+  std::unique_ptr<RuntimeContext> context(static_cast<RuntimeContext*>(data));
+  if (!context->pending) return;
+  try { closeRuntime(*context->pending); }
   catch (const Failure& failure) {
     // Environment teardown cannot retain an owner or return an error. Report
     // terminal cleanup failure instead of claiming the directory was removed.
-    std::fprintf(stderr, "REA Windows probe cleanup failed (Win32 %lu) path=\"",
+    std::fprintf(stderr, "REA Windows runtime cleanup failed (Win32 %lu) path=\"",
                  static_cast<unsigned long>(failure.win32));
     // Emit reversible UTF-16 escapes without allocation or CRT locale conversion.
     for (wchar_t unit : failure.path) {
@@ -112,10 +131,10 @@ static void cleanupInspection(void* data) noexcept {
     std::fprintf(stderr, "\"\n");
   }
   catch (const std::exception& failure) {
-    std::fprintf(stderr, "REA Windows probe cleanup failed: %s\n", failure.what());
+    std::fprintf(stderr, "REA Windows runtime cleanup failed: %s\n", failure.what());
   }
   catch (...) {
-    std::fprintf(stderr, "REA Windows probe cleanup failed: unknown native exception\n");
+    std::fprintf(stderr, "REA Windows runtime cleanup failed: unknown native exception\n");
   }
 }
 
@@ -128,12 +147,13 @@ static napi_value call(napi_env env, napi_callback_info info) {
     require(count == 2, "Native call requires operation and arguments", L"", ERROR_INVALID_PARAMETER);
     const auto operation = wide(env, args[0]);
     const auto values = array(env, args[1]);
+    auto& context = *static_cast<RuntimeContext*>(data);
     if (operation == L"inspect") {
       require(values.empty(), "Inspect takes no arguments", L"", ERROR_INVALID_PARAMETER);
-      return inspect(env, static_cast<InspectionContext*>(data)->probe);
+      return inspect(env, context.pending);
     }
     if (operation.rfind(L"process_", 0) == 0) return processCall(env, operation, values);
-    return filesystemCall(env, operation, values);
+    return filesystemCall(env, operation, values, context);
   } catch (const Failure& failure) {
     auto error = failureValue(env, failure);
     napi_throw(env, error);
@@ -145,8 +165,8 @@ static napi_value call(napi_env env, napi_callback_info info) {
 }
 
 static napi_value initialize(napi_env env, napi_value exports) {
-  auto context = std::make_unique<InspectionContext>();
-  if (napi_add_env_cleanup_hook(env, cleanupInspection, context.get()) != napi_ok) return nullptr;
+  auto context = std::make_unique<RuntimeContext>();
+  if (napi_add_env_cleanup_hook(env, cleanupRuntimeContext, context.get()) != napi_ok) return nullptr;
   // The environment owns callback data before a function can expose it. Failed
   // initialization still leaves a valid context for the registered cleanup hook.
   auto* data = context.release();

native/windows/src/authority.hpp

@@ -33,6 +33,9 @@ struct Failure : std::runtime_error {
   DWORD win32;
   std::wstring path;
   std::string constraint;
+  std::string cleanupConstraint;
+  DWORD cleanupWin32 = ERROR_SUCCESS;
+  std::wstring cleanupPath;
   // Affirmative proof for a failed runtime-parent admission; omission is unknown.
   bool runtimeDirectoryUnallocated = false;
   Failure(std::string constraint, std::wstring path = L"", DWORD code = GetLastError())
@@ -96,6 +99,9 @@ struct Runtime : Resource {
   std::atomic_bool snapshotCancelled{false};
   Runtime() : Resource(Kind::Runtime) {}
 };
+// A failed, not-yet-published runtime remains owned by its N-API environment
+// until a later admission or environment teardown can retry cleanup by handle.
+struct RuntimeContext { std::unique_ptr<Runtime> pending; };
 struct Process : Resource {
   Handle job, process, stdoutRead, stderrRead, stdinWrite;
   // Present only for a job with a commit limit; receives the job's notifications.
@@ -120,21 +126,24 @@ std::wstring wide(napi_env env, napi_value value);
 double numeric(napi_env env, napi_value value);
 std::vector<napi_value> array(napi_env env, napi_value value);
 napi_value wrap(napi_env env, std::unique_ptr<Resource> resource);
+void publishRuntimeHandle(napi_env env, napi_value observation, std::unique_ptr<Runtime>& runtime);
 Resource& resource(napi_env env, napi_value value, Kind kind);
 
 std::unique_ptr<File> openFile(const std::wstring& path, DWORD access = GENERIC_READ,
                              bool directory = false,
                              DWORD finalSharing = FILE_SHARE_READ);
 napi_value identity(napi_env env, HANDLE handle, const std::wstring& path);
-std::unique_ptr<Runtime> createRuntime(const std::wstring& parent, const std::wstring& prefix);
+void createRuntime(const std::wstring& parent, const std::wstring& prefix,
+                   std::unique_ptr<Runtime>& owner);
+void cleanupFailedRuntime(std::unique_ptr<Runtime>& owner, Failure& primary);
 void closeRuntime(Runtime& runtime);
 napi_value filesystemCall(napi_env env, const std::wstring& operation,
-                          const std::vector<napi_value>& args);
+                          const std::vector<napi_value>& args, RuntimeContext& context);
 std::unique_ptr<Process> spawnProcess(const std::wstring& command, const std::wstring& commandLine,
                                     const std::wstring& cwd, const std::vector<std::wstring>& environment,
                                     bool protocolStdin = false, ULONGLONG memoryLimit = 0);
 napi_value processCall(napi_env env, const std::wstring& operation,
                        const std::vector<napi_value>& args);
-napi_value inspect(napi_env env, std::unique_ptr<Runtime>& probe);
+napi_value inspect(napi_env env, std::unique_ptr<Runtime>& pending);
 
 } // namespace rea

native/windows/src/filesystem.cc

@@ -67,7 +67,10 @@ std::unique_ptr<File> openFile(const std::wstring& requested, DWORD access, bool
   result->requestedPath = requested;
   const auto& path = result->path;
   require(directory || path.size() > 3, "Target is not a regular file", requested, ERROR_DIRECTORY);
-  result->handles.push_back(openComponent(path.substr(0, 3), 0, true));
+  // The volume root cannot be replaced or deleted. Match the later root
+  // reopen's write sharing here so a root writer cannot block admission.
+  result->handles.push_back(openComponent(path.substr(0, 3), 0, true,
+                                          FILE_SHARE_READ | FILE_SHARE_WRITE));
   wchar_t filesystem[64]{};
   DWORD flags = 0;
   require(GetVolumeInformationByHandleW(result->handles.front().get(), nullptr, 0, nullptr, nullptr,
@@ -181,6 +184,29 @@ static Handle createRelative(HANDLE parent, const std::wstring& name, DWORD acce
   return result;
 }
 
+void cleanupFailedRuntime(std::unique_ptr<Runtime>& owner, Failure& primary) {
+  if (!owner || !owner->directory.valid()) {
+    owner.reset();
+    return;
+  }
+  try {
+    closeRuntime(*owner);
+    owner.reset();
+  } catch (const Failure& cleanup) {
+    primary.cleanupConstraint = cleanup.constraint;
+    primary.cleanupWin32 = cleanup.win32;
+    primary.cleanupPath = cleanup.path;
+  } catch (const std::exception& cleanup) {
+    primary.cleanupConstraint = cleanup.what();
+    primary.cleanupWin32 = ERROR_GEN_FAILURE;
+    primary.cleanupPath = owner->path;
+  } catch (...) {
+    primary.cleanupConstraint = "Unknown runtime cleanup failure";
+    primary.cleanupWin32 = ERROR_GEN_FAILURE;
+    primary.cleanupPath = owner->path;
+  }
+}
+
 // Lookup relative to an already-owned directory; path is diagnostic only.
 static Handle openRelative(HANDLE parent, const std::wstring& name, DWORD access,
                            DWORD sharing, ULONG kind, const std::wstring& path) {
@@ -334,7 +360,11 @@ static napi_value readCutterBridgeDescriptor(napi_env env, const std::vector<nap
   return result;
 }
 
-std::unique_ptr<Runtime> createRuntime(const std::wstring& parentPath, const std::wstring& prefix) {
+void createRuntime(const std::wstring& parentPath, const std::wstring& prefix,
+                   std::unique_ptr<Runtime>& owner) {
+  // A previous failed admission retains the original root handle. Retry that
+  // owner before allowing this environment to allocate another runtime.
+  if (owner) { closeRuntime(*owner); owner.reset(); }
   std::unique_ptr<File> parent;
   try { parent = openFile(parentPath, 0, true); }
   catch (Failure& failure) {
@@ -346,24 +376,29 @@ std::unique_ptr<Runtime> createRuntime(const std::wstring& parentPath, const std
   std::array<unsigned char, 16> random;
   require(BCryptGenRandom(nullptr, random.data(), random.size(), BCRYPT_USE_SYSTEM_PREFERRED_RNG) == 0,
           "Generate runtime identity failed", parentPath, ERROR_GEN_FAILURE);
+  Security security;
   auto result = std::make_unique<Runtime>();
   const auto name = prefix + hex(random.data(), random.size());
   result->path = parent->path + (parent->path.back() == L'\\' ? L"" : L"\\") + name;
-  Security security;
-  result->directory = createRelative(parent->get(), name,
-                                     DELETE | FILE_LIST_DIRECTORY | FILE_READ_ATTRIBUTES | READ_CONTROL,
-                                     true, security, result->path);
+  result->parents = std::move(parent->handles);
+  owner = std::move(result);
   try {
-    verifyPrivate(result->directory.get(), result->path, true);
-    // The owned child pins the parent nonempty before write sharing is relaxed.
-    parent->handles.back() = openComponent(parent->path, 0, true, FILE_SHARE_READ | FILE_SHARE_WRITE);
-    result->parents = std::move(parent->handles);
-  } catch (...) {
-    FILE_DISPOSITION_INFO deletion{TRUE};
-    SetFileInformationByHandle(result->directory.get(), FileDispositionInfo, &deletion, sizeof(deletion));
+    owner->directory = createRelative(owner->parents.back().get(), name,
+                                      DELETE | FILE_LIST_DIRECTORY | FILE_READ_ATTRIBUTES | FILE_WRITE_ATTRIBUTES | READ_CONTROL,
+                                      true, security, owner->path);
+    // FILE_CREATE proves ownership and pins the parent nonempty independently
+    // of admission validation. A failed validation must not retain a strict
+    // parent lease that prevents renaming the still-owned child before retry.
+    owner->parents.back() = openComponent(parent->path, 0, true, FILE_SHARE_READ | FILE_SHARE_WRITE);
+    verifyPrivate(owner->directory.get(), owner->path, true);
+  } catch (Failure& primary) {
+    cleanupFailedRuntime(owner, primary);
     throw;
+  } catch (const std::exception& error) {
+    Failure primary(error.what(), owner->path, ERROR_GEN_FAILURE);
+    cleanupFailedRuntime(owner, primary);
+    throw primary;
   }
-  return result;
 }
 
 static std::wstring relativePath(Runtime& root, const std::wstring& relative) {
@@ -438,6 +473,9 @@ static void dispose(HANDLE handle, const std::wstring& path) {
   require(GetFileInformationByHandleEx(handle, FileBasicInfo, &info, sizeof(info)), "Read cleanup attributes failed", path);
   if ((info.FileAttributes & FILE_ATTRIBUTE_READONLY) != 0) {
     info.FileAttributes &= ~FILE_ATTRIBUTE_READONLY;
+    // Zero means "leave attributes unchanged" to FileBasicInfo. A file whose
+    // sole attribute was READONLY needs NORMAL to actually clear that marker.
+    if (info.FileAttributes == 0) info.FileAttributes = FILE_ATTRIBUTE_NORMAL;
     require(SetFileInformationByHandle(handle, FileBasicInfo, &info, sizeof(info)), "Clear owned readonly attribute failed", path);
   }
   FILE_DISPOSITION_INFO deletion{TRUE};
@@ -654,7 +692,8 @@ static napi_value snapshot(napi_env env, napi_value rootValue, Runtime& root,
   return promise;
 }
 
-napi_value filesystemCall(napi_env env, const std::wstring& operation, const std::vector<napi_value>& args) {
+napi_value filesystemCall(napi_env env, const std::wstring& operation,
+                          const std::vector<napi_value>& args, RuntimeContext& context) {
   if (operation == L"cutter_bridge_verify_directory") return verifyCutterBridgeDirectory(env, args);
   if (operation == L"cutter_bridge_read_descriptor") return readCutterBridgeDescriptor(env, args);
   const auto expected = operation == L"open" || operation == L"close" || operation == L"runtime_close" || operation == L"runtime_snapshot_cancel" ? 1
@@ -688,12 +727,21 @@ napi_value filesystemCall(napi_env env, const std::wstring& operation, const std
     return result;
   }
   if (operation == L"runtime_create") {
-    auto root = createRuntime(wide(env, args[0]), wide(env, args[1]));
-    auto result = identity(env, root->directory.get(), root->path);
-    set(env, result, "path", string(env, root->path));
-    set(env, result, "privateDacl", boolean(env, true));
-    set(env, result, "handle", wrap(env, std::move(root)));
-    return result;
+    createRuntime(wide(env, args[0]), wide(env, args[1]), context.pending);
+    try {
+      auto result = identity(env, context.pending->directory.get(), context.pending->path);
+      set(env, result, "path", string(env, context.pending->path));
+      set(env, result, "privateDacl", boolean(env, true));
+      publishRuntimeHandle(env, result, context.pending);
+      return result;
+    } catch (Failure& primary) {
+      cleanupFailedRuntime(context.pending, primary);
+      throw;
+    } catch (const std::exception& error) {
+      Failure primary(error.what(), context.pending ? context.pending->path : L"", ERROR_GEN_FAILURE);
+      cleanupFailedRuntime(context.pending, primary);
+      throw primary;
+    }
   }
   auto& root = static_cast<Runtime&>(resource(env, args[0], Kind::Runtime));
   require(!root.closing || operation == L"runtime_close", "Runtime cleanup is pending", root.path, ERROR_BUSY);

native/windows/src/process.cc

@@ -454,17 +454,27 @@ napi_value processCall(napi_env env, const std::wstring& operation, const std::v
   throw Failure("Unknown native process operation", L"", ERROR_INVALID_PARAMETER);
 }
 
-napi_value inspect(napi_env env, std::unique_ptr<Runtime>& probe) {
+napi_value inspect(napi_env env, std::unique_ptr<Runtime>& pending) {
   // Settle the previous owner before admitting another probe. Failed cleanup
   // retains its handles, including when identity construction itself failed.
-  if (probe) { closeRuntime(*probe); probe.reset(); }
+  if (pending) { closeRuntime(*pending); pending.reset(); }
   wchar_t temp[32768];
   const auto length = GetTempPathW(32768, temp);
   require(length > 0 && length < 32768, "Read temporary directory coordinate failed");
-  probe = createRuntime(std::wstring(temp, length), L"rea-native-probe-");
-  auto filesystem = identity(env, probe->directory.get(), probe->path);
-  closeRuntime(*probe);
-  probe.reset();
+  createRuntime(std::wstring(temp, length), L"rea-native-probe-", pending);
+  napi_value filesystem;
+  try {
+    filesystem = identity(env, pending->directory.get(), pending->path);
+  } catch (Failure& primary) {
+    cleanupFailedRuntime(pending, primary);
+    throw;
+  } catch (const std::exception& error) {
+    Failure primary(error.what(), pending ? pending->path : L"", ERROR_GEN_FAILURE);
+    cleanupFailedRuntime(pending, primary);
+    throw primary;
+  }
+  closeRuntime(*pending);
+  pending.reset();
   auto job = createJob();
   Attributes attributes(1);
   HANDLE raw = job.get();

package-lock.json

@@ -34,6 +34,7 @@
         "isomorphic-git": "1.38.7",
         "jsonc-parser": "3.3.1",
         "lossless-json": "4.3.1",
+        "openpgp": "6.3.2",
         "parse5": "8.0.1",
         "pino": "10.4.0",
         "playwright-core": "1.58.2",
@@ -7871,6 +7872,24 @@
         "regex-recursion": "^6.0.2"
       }
     },
+    "node_modules/openpgp": {
+      "version": "6.3.2",
+      "resolved": "https://registry.npmjs.org/openpgp/-/openpgp-6.3.2.tgz",
+      "integrity": "sha512-wcZTzHz41LV8Y48zH/JlD1JT8YdNmpWOuMAjQ/podvvCwsjBEU37K3znJh4UQcj3hksE5z9TzvhbUbwzsGvlLQ==",
+      "license": "LGPL-3.0+",
+      "engines": {
+        "node": ">= 18.0.0",
+        "typescript": ">= 5.0.0"
+      },
+      "peerDependencies": {
+        "@openpgp/web-stream-tools": "~0.3.0"
+      },
+      "peerDependenciesMeta": {
+        "@openpgp/web-stream-tools": {
+          "optional": true
+        }
+      }
+    },
     "node_modules/oxc-parser": {
       "version": "0.150.0",
       "resolved": "https://registry.npmjs.org/oxc-parser/-/oxc-parser-0.150.0.tgz",

package.json

@@ -337,6 +337,7 @@
     "isomorphic-git": "1.38.7",
     "jsonc-parser": "3.3.1",
     "lossless-json": "4.3.1",
+    "openpgp": "6.3.2",
     "parse5": "8.0.1",
     "pino": "10.4.0",
     "playwright-core": "1.58.2",

scripts/ci/scopes.mjs

@@ -138,6 +138,7 @@ export const SCOPES = {
     "src/domain/javascript/javascriptRuntimeStaticCandidates.ts",
     "tests/boundary/filesystem/webScriptManifestPaths.test.ts",
     "tests/boundary/process/ProcessEvidence.host.test.ts",
+    "tests/boundary/process/windowsOutputCapture.test.ts",
     "tests/acceptance/webModuleCli.test.ts",
     "tests/acceptance/webSourceLocationCli.test.ts",
     "tests/boundary/cli/referenceSourceImportStatus.test.ts",

scripts/create-ghidra-windows-fixture.mjs

@@ -75,7 +75,57 @@ for (const architecture of ["x86_64", "x86"]) {
         : [0xe8, 0x0b, 0, 0, 0, 0xe8, 0x16, 0, 0, 0, 0x31, 0xc0, 0xc3],
       0x200,
     );
-    bytes.set([0xb8, 0x2a, 0, 0, 0, 0xc3], 0x210);
+    bytes.set(
+      x86 && !dll
+        ? [0xb8, 0x2a, 0, 0, 0, 0xe8, 0x66, 0, 0, 0, 0xc3]
+        : [0xb8, 0x2a, 0, 0, 0, 0xc3],
+      0x210,
+    );
+
+    if (x86 && !dll) {
+      // A statically analyzed Borland-style SEH registration whose handler
+      // entry is referenced but is not part of the registering function body.
+      bytes.set(
+        [
+          0x55, // PUSH EBP
+          0x89,
+          0xe5, // MOV EBP, ESP
+          0x68,
+          0xc0,
+          0x10,
+          0x40,
+          0x00, // PUSH 0x4010c0
+          0x64,
+          0xff,
+          0x35,
+          0,
+          0,
+          0,
+          0, // PUSH DWORD PTR FS:[0]
+          0x64,
+          0x89,
+          0x25,
+          0,
+          0,
+          0,
+          0, // MOV DWORD PTR FS:[0], ESP
+          0x64,
+          0x8f,
+          0x05,
+          0,
+          0,
+          0,
+          0, // POP DWORD PTR FS:[0]
+          0x83,
+          0xc4,
+          0x04, // ADD ESP, 4
+          0x5d, // POP EBP
+          0xc3, // RET
+        ],
+        0x280,
+      );
+      bytes.set([0xc3], 0x2c0);
+    }
 
     if (!dll) {
       const dataSection = sectionTable + 40;

scripts/hopper-demo-x11.py

@@ -1,5 +1,5 @@
 #!/usr/bin/env python3
-"""Inspect or activate Hopper's version-pinned demo dialog on a private X display."""
+"""Inspect or activate Hopper's authenticated demo dialog on a private X display."""
 
 from __future__ import annotations
 
@@ -8,6 +8,7 @@ import ctypes
 import hashlib
 import json
 import os
+import re
 from pathlib import Path
 import secrets
 import select
@@ -18,12 +19,6 @@ import threading
 import time
 
 
-# Keep this digest aligned with LinuxHopper.ts and rerun real-Hopper verification
-# before accepting another vendor build.
-SUPPORTED_HOPPER_SHA256 = {
-    "0294ced141cc373468ee22d8343e7dac41980cb05a937994ca81c9f09afe7ded",
-    "1339f9e58377442b0c6fcb0dfc3cec20d593cc557408521fad9a00dbc6b8da13",
-}
 EXPECTED_SCREEN = (1280, 1024)
 DIALOG_TITLES = {"Registration", "Registration — Hopper Disassembler"}
 HOPPER_WINDOW_TITLE = "Hopper Disassembler"
@@ -121,6 +116,7 @@ def parse_args() -> argparse.Namespace:
     parser.add_argument("--probe", action="store_true")
     parser.add_argument("--owner-stdin", action="store_true")
     parser.add_argument("--hopper")
+    parser.add_argument("--expected-launcher-sha256")
     parser.add_argument("--socket")
     parser.add_argument(
         "--strategy",
@@ -661,7 +657,11 @@ def run_probe(args: argparse.Namespace) -> dict[str, object]:
 
 
 def run_launch(args: argparse.Namespace) -> int:
-    if args.hopper is None or args.socket is None:
+    if (
+        args.hopper is None or args.socket is None
+        or args.expected_launcher_sha256 is None
+        or re.fullmatch(r"[a-f0-9]{64}", args.expected_launcher_sha256) is None
+    ):
         raise AdapterFailure(
             INVALID_LAUNCH_COMMAND,
             "invalid_launch_command",
@@ -673,7 +673,9 @@ def run_launch(args: argparse.Namespace) -> int:
     hopper = Path(args.hopper)
     socket = Path(args.socket)
     try:
-        supported = digest(hopper) in SUPPORTED_HOPPER_SHA256
+        # The parent owns publisher/package interpretation. Recheck its admitted
+        # launcher bytes immediately before execution, preserving all X11 guards.
+        supported = digest(hopper) == args.expected_launcher_sha256
     except OSError as error:
         raise AdapterFailure(
             INVALID_LAUNCH_COMMAND,
@@ -684,7 +686,7 @@ def run_launch(args: argparse.Namespace) -> int:
         raise AdapterFailure(
             UNSUPPORTED_HOPPER_BUILD,
             "unsupported_hopper_build",
-            "unsupported Hopper binary",
+            "Hopper binary differs from the authenticated release",
         )
     command = args.command[1:] if args.command[:1] == ["--"] else args.command
     if not command or Path(command[0]).resolve() != hopper.resolve():

scripts/lib/browser-scenario-environment-e2e.mjs

@@ -4,9 +4,11 @@ import { fileURLToPath } from "node:url";
 import { Client } from "@modelcontextprotocol/client";
 import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
 
+import { CdpCommandRejection } from "../../dist/browser/CdpCommandRejection.js";
 import { CdpConnection } from "../../dist/browser/CdpConnection.js";
 import { createBrowserScenarioProvider } from "../../dist/composition/browserScenario.js";
 import { browserScenarioSchema } from "../../dist/domain/browserScenario.js";
+import { AnalysisCancelledError } from "../../dist/domain/analysisErrorCore.js";
 import { runScenarioCli } from "./browser-scenario-verifier.mjs";
 import { requireMcpEvidenceResult } from "./mcp-verifier-results.mjs";
 
@@ -24,12 +26,24 @@ const alternate = {
   width: 700,
   height: 500,
 };
-const expression = `JSON.stringify({
+const environmentExpression = `{
   timezone: Intl.DateTimeFormat().resolvedOptions().timeZone,
   locale: Intl.DateTimeFormat().resolvedOptions().locale,
   dpr: devicePixelRatio, width: innerWidth, height: innerHeight
+}`;
+const expression = `JSON.stringify(${environmentExpression})`;
+const scenarioSnapshotExpression = `JSON.stringify({
+  timeOrigin: performance.timeOrigin,
+  environment: ${environmentExpression},
+  run: document.body?.dataset.run ?? null
 })`;
 
+const isNavigationTransitionRejection = (cause) =>
+  cause instanceof CdpCommandRejection &&
+  cause.command === "Runtime.evaluate" &&
+  cause.code === -32_000 &&
+  cause.reportedMessage === "Inspected target navigated or closed";
+
 const environmentScenario = ({ endpoint, targetId, origin, values, run }) =>
   browserScenarioSchema.parse({
     browser: { mode: "connect", cdp_endpoint: endpoint, target_id: targetId },
@@ -146,7 +160,6 @@ export async function verifyScenarioEnvironment(endpoint, targetId, origin) {
         endpoint,
         targetId,
         origin,
-        readEnvironment,
         assertRestored,
       });
     } finally {
@@ -178,12 +191,7 @@ export async function verifyScenarioEnvironment(endpoint, targetId, origin) {
     assertCapture(actionFailure.value);
     assert.equal(actionFailure.value.steps.at(-1).status, "failed");
     await assertRestored();
-    await verifyCancelledEnvironment(
-      provider,
-      connection,
-      waiting,
-      readEnvironment,
-    );
+    await verifyCancelledEnvironment(provider, connection, waiting);
     await assertRestored();
     return {
       mocked: false,
@@ -208,7 +216,6 @@ const verifyConcurrentMcpCapture = async ({
   endpoint,
   targetId,
   origin,
-  readEnvironment,
   assertRestored,
 }) => {
   const firstScenario = environmentScenario({
@@ -241,12 +248,13 @@ const verifyConcurrentMcpCapture = async ({
 
   try {
     const first = callTool(firstScenario, firstCancellation.signal);
-    await waitForEnvironment(readEnvironment, connection, expected, "A");
+    await waitForEnvironment(connection, expected, "A");
 
     const second = callTool(secondScenario, secondCancellation.signal);
     await client.ping();
-    assert.deepEqual(await readEnvironment(), expected);
-    assert.equal(await readScenarioRun(connection), "A");
+    const firstSnapshot = await readScenarioSnapshot(connection);
+    assert.deepEqual(firstSnapshot.environment, expected);
+    assert.equal(firstSnapshot.run, "A");
 
     const cancelled = callTool(
       environmentScenario({
@@ -261,8 +269,9 @@ const verifyConcurrentMcpCapture = async ({
     await client.ping();
     cancellation.abort(new Error("cancel queued browser scenario"));
     await assert.rejects(cancelled, /cancel queued browser scenario/u);
-    assert.deepEqual(await readEnvironment(), expected);
-    assert.equal(await readScenarioRun(connection), "A");
+    const queuedSnapshot = await readScenarioSnapshot(connection);
+    assert.deepEqual(queuedSnapshot.environment, expected);
+    assert.equal(queuedSnapshot.run, "A");
 
     await releaseScenario(connection);
     assertCapture(
@@ -270,7 +279,7 @@ const verifyConcurrentMcpCapture = async ({
       expected,
     );
 
-    await waitForEnvironment(readEnvironment, connection, alternate, "B");
+    await waitForEnvironment(connection, alternate, "B");
     await releaseScenario(connection);
     assertCapture(
       requireMcpEvidenceResult(await second, "capture_browser_scenario"),
@@ -287,31 +296,57 @@ const verifyConcurrentMcpCapture = async ({
   }
 };
 
-const waitForEnvironment = async (
-  readEnvironment,
-  connection,
-  expectedValues,
-  run,
-) => {
-  const deadline = Date.now() + 10_000;
-  while (Date.now() < deadline) {
-    if (
-      JSON.stringify(await readEnvironment()) ===
-        JSON.stringify(expectedValues) &&
-      (await readScenarioRun(connection)) === run
-    )
-      return;
-    await delay(25);
-  }
-  throw new Error(`Scenario ${run} never reached its page-controlled gate`);
+const waitForEnvironment = async (connection, expectedValues, run) => {
+  await waitForScenarioSnapshot(
+    connection,
+    (snapshot) =>
+      JSON.stringify(snapshot.environment) === JSON.stringify(expectedValues) &&
+      snapshot.run === run,
+    `Scenario ${run} never reached its page-controlled gate`,
+  );
 };
 
-const readScenarioRun = async (connection) => {
-  const value = await connection.send("Runtime.evaluate", {
-    expression: "document.body.dataset.run ?? null",
-    returnByValue: true,
-  });
-  return value.result.value;
+const waitForScenarioSnapshot = async (connection, accept, message) => {
+  const controller = new AbortController();
+  const timer = setTimeout(() => controller.abort(), 10_000);
+  let lastTransitionRejection;
+  try {
+    while (!controller.signal.aborted) {
+      try {
+        const snapshot = await readScenarioSnapshot(
+          connection,
+          controller.signal,
+        );
+        if (controller.signal.aborted) break;
+        if (accept(snapshot)) return;
+      } catch (cause) {
+        if (
+          controller.signal.aborted &&
+          cause instanceof AnalysisCancelledError
+        )
+          break;
+        if (!isNavigationTransitionRejection(cause)) throw cause;
+        lastTransitionRejection = cause;
+      }
+      await delay(25);
+    }
+    throw new Error(message, {
+      cause: lastTransitionRejection ?? controller.signal.reason,
+    });
+  } finally {
+    clearTimeout(timer);
+  }
+};
+
+const readScenarioSnapshot = async (connection, signal) => {
+  const value = await connection.send(
+    "Runtime.evaluate",
+    { expression: scenarioSnapshotExpression, returnByValue: true },
+    undefined,
+    signal,
+  );
+  assert.equal(value.result.type, "string");
+  return JSON.parse(value.result.value);
 };
 
 const releaseScenario = async (connection) => {
@@ -321,39 +356,26 @@ const releaseScenario = async (connection) => {
   });
 };
 
-async function verifyCancelledEnvironment(
-  provider,
-  connection,
-  scenario,
-  readEnvironment,
-) {
+async function verifyCancelledEnvironment(provider, connection, scenario) {
   const controller = new AbortController();
-  const priorNavigation = await connection.send("Runtime.evaluate", {
-    expression: "performance.timeOrigin",
-    returnByValue: true,
-  });
+  const priorSnapshot = await readScenarioSnapshot(connection);
   const pending = provider.captureScenario(
     { ...scenario, actions: [{ ...scenario.actions[0], timeout_ms: 30_000 }] },
     { signal: controller.signal },
   );
+  const settlement = Promise.allSettled([pending]);
   try {
-    const deadline = Date.now() + 10_000;
-    while (true) {
-      const navigation = await connection.send("Runtime.evaluate", {
-        expression: "performance.timeOrigin",
-        returnByValue: true,
-      });
-      if (
-        navigation.result.value !== priorNavigation.result.value &&
-        JSON.stringify(await readEnvironment()) === JSON.stringify(expected)
-      )
-        break;
-      assert.ok(Date.now() < deadline, "Scenario never applied emulation");
-      await delay(25);
-    }
+    await waitForScenarioSnapshot(
+      connection,
+      (snapshot) =>
+        snapshot.timeOrigin !== priorSnapshot.timeOrigin &&
+        JSON.stringify(snapshot.environment) === JSON.stringify(expected),
+      "Scenario never applied emulation",
+    );
     await delay(100);
   } finally {
     controller.abort();
+    await settlement;
   }
   const cancelled = await pending;
   assert.ok(

scripts/lib/browser-scenario-service-workers-e2e.mjs

@@ -144,14 +144,19 @@ export async function verifyScenarioServiceWorkers(endpoint, targetId) {
         true,
       );
     };
-    const assertRejected = async (scenario) => {
+    const assertRejected = async (scenario, expectedReason) => {
       const result = await provider.captureScenario(scenario);
       assert.equal(
         result.ok,
         false,
         "Worker-influenced capture unexpectedly succeeded",
       );
-      assert.equal(result.error.reason, "target_not_allowed");
+      assert.equal(result.error.reason, expectedReason);
+      if (expectedReason === "service_worker_blocked")
+        assert.match(
+          result.error.userMessage,
+          /environment\.service_workers to allow/,
+        );
       assert.equal(
         result.error.partialObservation.capture.browser.cleanup,
         "disconnected-external",
@@ -229,7 +234,7 @@ export async function verifyScenarioServiceWorkers(endpoint, targetId) {
       "worker",
     );
     await page.goto("about:blank");
-    await assertRejected(scenarioFor(second.origin));
+    await assertRejected(scenarioFor(second.origin), "service_worker_blocked");
     await page.goto("about:blank");
     await assertRejected(
       scenarioFor(first.origin, [
@@ -240,6 +245,7 @@ export async function verifyScenarioServiceWorkers(endpoint, targetId) {
           wait_until: "load",
         },
       ]),
+      "service_worker_blocked",
     );
     await page.goto("about:blank");
     await assertRejected(
@@ -251,9 +257,13 @@ export async function verifyScenarioServiceWorkers(endpoint, targetId) {
         },
         { step_id: "loaded", action: "wait_for_timeout", duration_ms: 100 },
       ]),
+      "service_worker_blocked",
     );
     await page.goto("about:blank");
-    await assertRejected(scenarioFor(`${first.origin}/frame`));
+    await assertRejected(
+      scenarioFor(`${first.origin}/frame`),
+      "service_worker_blocked",
+    );
 
     const allowed = await provider.captureScenario(
       scenarioFor(second.origin, [], "allow"),

scripts/lib/ghidra-verifier-timeouts.d.mts

@@ -0,0 +1,33 @@
+export interface GhidraVerifierTimeouts {
+  readonly clientOptions: {
+    readonly startupTimeoutMs: number;
+  };
+  readonly requestOptions: {
+    readonly timeout: number;
+  };
+}
+
+/** Resolve one verifier-owned timeout policy before target I/O or allocation. */
+export function resolveGhidraVerifierTimeouts(
+  environment: Readonly<Record<string, string | undefined>>,
+): GhidraVerifierTimeouts;
+
+export interface GhidraVerifierMcpClient {
+  connect(
+    transport: unknown,
+    options: GhidraVerifierTimeouts["requestOptions"],
+  ): Promise<void>;
+  callTool(
+    request: unknown,
+    options: GhidraVerifierTimeouts["requestOptions"],
+  ): Promise<unknown>;
+}
+
+/** Bind every verifier MCP operation to the same owned request budget. */
+export function createGhidraVerifierMcpOperations(
+  client: GhidraVerifierMcpClient,
+  requestOptions: GhidraVerifierTimeouts["requestOptions"],
+): {
+  connect(transport: unknown): Promise<void>;
+  callTool(request: unknown): Promise<unknown>;
+};

scripts/lib/ghidra-verifier-timeouts.mjs

@@ -0,0 +1,31 @@
+import { parseConfig } from "../../dist/config/parseConfig.js";
+
+const FILE_OFFSET_OPERATION_TIMEOUT_MS = 240_000;
+const MAX_TIMER_DELAY_MS = 2_147_483_647;
+
+/** Resolve one verifier-owned timeout policy before target I/O or allocation. */
+export function resolveGhidraVerifierTimeouts(environment) {
+  const configuration = parseConfig(environment);
+  if (!configuration.ok) throw configuration.error;
+  const startupTimeoutMs = configuration.value.ghidraStartupTimeoutMs;
+  // A cold public query includes provider startup before its analysis request.
+  // Preserve the operation budget after the caller-selected startup deadline.
+  const publicRequestTimeoutMs =
+    startupTimeoutMs + FILE_OFFSET_OPERATION_TIMEOUT_MS;
+  if (publicRequestTimeoutMs > MAX_TIMER_DELAY_MS)
+    throw new RangeError(
+      "The Ghidra startup and verification operation budgets exceed the Node.js timer limit.",
+    );
+  return {
+    clientOptions: { startupTimeoutMs },
+    requestOptions: { timeout: publicRequestTimeoutMs },
+  };
+}
+
+/** Bind every verifier MCP operation to the same owned request budget. */
+export function createGhidraVerifierMcpOperations(client, requestOptions) {
+  return {
+    connect: (transport) => client.connect(transport, requestOptions),
+    callTool: (request) => client.callTool(request, requestOptions),
+  };
+}

scripts/lib/real-ghidra-borland-seh.mjs

@@ -0,0 +1,84 @@
+import assert from "node:assert/strict";
+
+const limitationPattern = /Borland\/Delphi-style SEH frame registration/u;
+
+/** Verify Ghidra reports, and does not absorb, the fixture's handler region. */
+export async function verifyGhidraBorlandSeh(functionObservation, x86) {
+  assert.equal(x86, true, "The SEH control exists only in the x86 PE fixture");
+  const caller = "0x401015";
+  const procedure = "0x401080";
+  const handler = "0x4010c0";
+
+  const ordinary = await functionObservation("procedure_info", {
+    procedure: "0x401010",
+  });
+  assert.ok(
+    !ordinary.limitations.some((item) => limitationPattern.test(item)),
+    "The ordinary return-42 control unexpectedly reports an SEH limitation",
+  );
+
+  const info = await functionObservation("procedure_info", { procedure });
+  assert.ok(info.limitations.some((item) => limitationPattern.test(item)));
+
+  const outgoing = await functionObservation("procedure_references", {
+    procedure,
+    direction: "outgoing",
+  });
+  assert.equal(outgoing.value.direction, "outgoing");
+  assert.ok(outgoing.limitations.some((item) => limitationPattern.test(item)));
+  assert.ok(
+    outgoing.value.references.some(
+      (reference) =>
+        reference.source_address === "0x401083" &&
+        reference.target_address === handler &&
+        reference.kind.available &&
+        reference.kind.data &&
+        !reference.kind.call,
+    ),
+    "The pushed handler address must remain an observed data reference, without inventing a call",
+  );
+
+  const incoming = await functionObservation("procedure_references", {
+    procedure,
+    direction: "incoming",
+  });
+  assert.equal(incoming.value.direction, "incoming");
+  assert.ok(incoming.limitations.some((item) => limitationPattern.test(item)));
+  assert.ok(
+    incoming.value.references.some(
+      (reference) =>
+        reference.source_address === caller &&
+        reference.target_address === procedure,
+    ),
+    "The caller-to-registration procedure reference was not observed",
+  );
+
+  const dossier = await functionObservation("analyze_function", {
+    procedure,
+  });
+  assert.ok(dossier.limitations.some((item) => limitationPattern.test(item)));
+  assert.ok(
+    dossier.value.limitations.some((item) => limitationPattern.test(item)),
+    "The normalized function dossier lost the SEH limitation",
+  );
+  assert.equal(dossier.value.procedure.body.available, true);
+  assert.ok(dossier.value.procedure.body.ranges.length > 0);
+  assert.ok(
+    !dossier.value.procedure.body.ranges.some(
+      (range) =>
+        BigInt(range.start) <= BigInt(handler) &&
+        BigInt(range.end) >= BigInt(handler),
+    ),
+    "The handler was incorrectly claimed as part of the procedure body",
+  );
+
+  return {
+    limitation: true,
+    ordinary_control: true,
+    procedure_references_incoming_and_outgoing: true,
+    handler_data_reference: true,
+    caller_reference: true,
+    dossier_limitation: true,
+    handler_body_not_claimed: true,
+  };
+}

scripts/lib/real-ghidra-file-offsets.d.mts

@@ -0,0 +1,16 @@
+export interface GhidraFileOffsetVerification {
+  readonly cli_mcp_parity: true;
+  readonly explicit_and_numeric_offsets: true;
+  readonly invalid_and_unmapped_offsets: true;
+  readonly mapped_offset_without_procedure: true;
+  readonly ambiguous_mappings: "verified" | "unavailable-single-section";
+  readonly source_unchanged: true;
+  readonly cleanup: "complete";
+}
+
+/** Exercise file coordinates through public CLI/MCP using a source-owned PE fixture. */
+export function verifyGhidraFileOffsets(
+  targetPath: string,
+  entrypoint: string,
+  procedure: string,
+): Promise<GhidraFileOffsetVerification>;

scripts/lib/real-ghidra-file-offsets.mjs

@@ -14,6 +14,10 @@ import { promisify } from "node:util";
 import { Client } from "@modelcontextprotocol/client";
 import { StdioClientTransport } from "@modelcontextprotocol/client/stdio";
 import { parseEvidence } from "../../dist/domain/evidence.js";
+import {
+  createGhidraVerifierMcpOperations,
+  resolveGhidraVerifierTimeouts,
+} from "./ghidra-verifier-timeouts.mjs";
 import {
   requireMcpOperationResult,
   requireMcpToolError,
@@ -26,17 +30,22 @@ export async function verifyGhidraFileOffsets(
   entrypoint,
   procedure,
 ) {
+  const environment = {
+    ...process.env,
+    REA_ANALYSIS_PROVIDER: "ghidra",
+    REA_LOG_LEVEL: "silent",
+  };
+  const { requestOptions } = resolveGhidraVerifierTimeouts(environment);
+  const timeoutMs = requestOptions.timeout;
   const original = await readFile(targetPath);
   const workspace = await mkdtemp(join(tmpdir(), "rea-file-offsets-"));
   const runtime = join(workspace, "runtime");
   await mkdir(runtime);
   const env = {
-    ...process.env,
+    ...environment,
     TEMP: runtime,
     TMP: runtime,
     TMPDIR: runtime,
-    REA_ANALYSIS_PROVIDER: "ghidra",
-    REA_LOG_LEVEL: "silent",
   };
   const client = new Client({ name: "rea-file-offsets", version: "1" });
   const transport = new StdioClientTransport({
@@ -45,21 +54,19 @@ export async function verifyGhidraFileOffsets(
     env,
     stderr: "pipe",
   });
+  const mcp = createGhidraVerifierMcpOperations(client, requestOptions);
   let connected = false;
   let complete = false;
   let primaryFailure;
   const call = async (name, arguments_ = {}) => {
-    const result = await client.callTool(
-      { name, arguments: arguments_ },
-      { timeout: 240_000 },
-    );
+    const result = await mcp.callTool({ name, arguments: arguments_ });
     const projected = requireMcpOperationResult(result, name);
     return ["open_binary", "binary_session", "close_binary"].includes(name)
       ? projected
       : (result.structuredContent ?? JSON.parse(mcpTextValue(result)));
   };
   try {
-    await client.connect(transport);
+    await mcp.connect(transport);
     connected = true;
     await call("open_binary", { path: targetPath, provider_id: "ghidra" });
     const session = await call("binary_session");
@@ -90,12 +97,12 @@ export async function verifyGhidraFileOffsets(
         ),
       );
     }
-    const invalid = await client.callTool({
+    const invalid = await mcp.callTool({
       name: "procedure_pseudo_code",
       arguments: { procedure: "file-offset:0x20000000000000" },
     });
     assert.equal(requireMcpToolError(invalid).code, "invalid_request");
-    const unmapped = await client.callTool({
+    const unmapped = await mcp.callTool({
       name: "procedure_pseudo_code",
       arguments: { procedure: "file-offset:0x100000" },
     });
@@ -122,7 +129,7 @@ export async function verifyGhidraFileOffsets(
     assert.equal(headerOffset.normalized_result.file_offset, 0);
     for (const selector of ["file-offset:0x0", "0x0"]) {
       const missing = requireMcpToolError(
-        await client.callTool({
+        await mcp.callTool({
           name: "procedure_pseudo_code",
           arguments: { procedure: selector },
         }),
@@ -147,7 +154,7 @@ export async function verifyGhidraFileOffsets(
             "ghidra",
             "--json",
           ],
-          { env, timeout: 240_000, maxBuffer: 16 * 1024 * 1024 },
+          { env, timeout: timeoutMs, maxBuffer: 16 * 1024 * 1024 },
         )
       ).stdout,
     );
@@ -164,7 +171,7 @@ export async function verifyGhidraFileOffsets(
         "ghidra",
         "--json",
       ],
-      { env, timeout: 240_000, maxBuffer: 16 * 1024 * 1024 },
+      { env, timeout: timeoutMs, maxBuffer: 16 * 1024 * 1024 },
     ).then(
       () => {
         throw new Error(
@@ -183,13 +190,10 @@ export async function verifyGhidraFileOffsets(
       const ambiguousPath = join(workspace, "ambiguous.exe");
       await writeFile(ambiguousPath, alias);
       await call("open_binary", { path: ambiguousPath, provider_id: "ghidra" });
-      const ambiguous = await client.callTool(
-        {
-          name: "procedure_pseudo_code",
-          arguments: { procedure: `file-offset:${coordinate}` },
-        },
-        { timeout: 240_000 },
-      );
+      const ambiguous = await mcp.callTool({
+        name: "procedure_pseudo_code",
+        arguments: { procedure: `file-offset:${coordinate}` },
+      });
       assert.ok(
         JSON.stringify(requireMcpToolError(ambiguous)).includes(
           "multiple loaded mappings",

scripts/lib/windows-volume-root-append-lease.mjs

@@ -0,0 +1,118 @@
+import assert from "node:assert/strict";
+import { spawn } from "node:child_process";
+import { once } from "node:events";
+import { createInterface } from "node:readline";
+import { resolve } from "node:path";
+
+const withDeadline = async (promise, milliseconds, description) => {
+  let timer;
+  try {
+    return await Promise.race([
+      promise,
+      new Promise((_, reject) => {
+        timer = setTimeout(
+          () =>
+            reject(new Error(`${description} exceeded ${milliseconds} ms.`)),
+          milliseconds,
+        );
+      }),
+    ]);
+  } finally {
+    clearTimeout(timer);
+  }
+};
+
+export async function verifyVolumeRootAppendLease({
+  fixturePath,
+  volumeRoot,
+  verifyAdmission,
+}) {
+  const lease = spawn(
+    resolve(fixturePath),
+    ["volume-root-append-lease", volumeRoot],
+    { stdio: ["pipe", "pipe", "pipe"], windowsHide: true },
+  );
+  const lines = createInterface({ input: lease.stdout });
+  let leaseStderr = "";
+  lease.stderr.setEncoding("utf8");
+  lease.stderr.on("data", (chunk) => {
+    if (leaseStderr.length < 8192) leaseStderr += chunk;
+  });
+
+  const closed = once(lease, "close").then(([code, signal]) => ({
+    code,
+    signal,
+  }));
+  const leaseLine = once(lines, "line").then(([line]) => ({ line }));
+  const leaseError = once(lease, "error").then(([error]) => {
+    throw error;
+  });
+  let verificationFailure;
+  try {
+    const ready = await withDeadline(
+      Promise.race([leaseLine, closed, leaseError]),
+      10_000,
+      "Volume-root append lease readiness",
+    );
+    assert.equal(
+      "line" in ready,
+      true,
+      `Volume-root append lease exited before readiness (code ${ready.code}, signal ${ready.signal}): ${leaseStderr}`,
+    );
+    const observation = JSON.parse(ready.line);
+    assert.equal(
+      observation.writeAccessGranted,
+      true,
+      `Windows test token cannot hold FILE_ADD_SUBDIRECTORY on ${volumeRoot} (Win32 ${observation.win32Code}).`,
+    );
+    await verifyAdmission();
+  } catch (error) {
+    verificationFailure = error;
+  }
+
+  lines.close();
+  if (
+    lease.exitCode === null &&
+    lease.signalCode === null &&
+    !lease.stdin.destroyed
+  )
+    lease.stdin.end("x");
+
+  let releaseFailure;
+  try {
+    const result = await withDeadline(
+      closed,
+      10_000,
+      "Volume-root append lease release",
+    ).catch(async (error) => {
+      lease.kill();
+      try {
+        return await withDeadline(
+          closed,
+          5_000,
+          "Volume-root append lease termination",
+        );
+      } catch (terminationError) {
+        throw new AggregateError(
+          [error, terminationError],
+          `Volume-root append lease could not be released: ${leaseStderr}`,
+        );
+      }
+    });
+    assert.equal(
+      result.code,
+      0,
+      `Volume-root append lease exited with code ${result.code}, signal ${result.signal}: ${leaseStderr}`,
+    );
+  } catch (error) {
+    releaseFailure = error;
+  }
+
+  if (verificationFailure !== undefined && releaseFailure !== undefined)
+    throw new AggregateError(
+      [verificationFailure, releaseFailure],
+      "Volume-root append lease verification and cleanup both failed.",
+    );
+  if (verificationFailure !== undefined) throw verificationFailure;
+  if (releaseFailure !== undefined) throw releaseFailure;
+}

scripts/rea.mjs

@@ -14,6 +14,10 @@ process.env.REA_PACKAGE_VERSION = packageJson.version;
 const isMcpMode =
   args.length === 1 && (args[0] === "--mcp" || args[0] === "mcp");
 const isMcpDoctorMode = args[0] === "mcp" && args[1] === "doctor";
+/** Exit status of a writer terminated by SIGPIPE (128 + 13). */
+const SIGPIPE_EXIT_STATUS = 141;
+// Set once the reader of CLI output has closed the pipe.
+let stdoutClosed = false;
 const packageRoot = resolve(dirname(fileURLToPath(import.meta.url)), "..");
 
 // Windows home lookup throws for an empty USERPROFILE. Dependencies resolve
@@ -55,13 +59,14 @@ if (!(await compiledRuntimeExists(runtimeFiles))) {
   const { runEntrypoint } = await import("../dist/mcp/main.js");
   await runEntrypoint();
 } else if (isMcpDoctorMode) {
+  guardClosedStdout();
   const { runProductionMcpDoctorCli } =
     await import("../dist/cli-runtime/mcpDoctor.js");
   const result = await runProductionMcpDoctorCli(args.slice(2), {
     dispatcherPath: fileURLToPath(import.meta.url),
     packageRoot,
   });
-  process.stdout.write(result.output);
+  writeStdout(result.output);
   process.exitCode = result.exitCode;
 } else {
   const { cliConfigurationError, createCli } =
@@ -69,36 +74,64 @@ if (!(await compiledRuntimeExists(runtimeFiles))) {
   const { createCliResultOutput } =
     await import("../dist/cli-runtime/cliResultOutput.js");
   const {
+    normalizeCliOutputArguments,
     renderCliOutputArgumentError,
     renderEmptyFilteredCliOutput,
     validateCliOutputArguments,
   } = await import("../dist/cli-runtime/cliOutput.js");
-  const outputArguments = validateCliOutputArguments(args);
+  guardClosedStdout();
+  const cliArgs = normalizeCliOutputArguments(args);
+  const outputArguments = validateCliOutputArguments(cliArgs);
   const configurationError = cliConfigurationError(process.env);
   if (configurationError !== undefined) {
     process.stderr.write(`${configurationError}\n`);
     process.exitCode = 1;
   } else if (!outputArguments.ok) {
-    process.stdout.write(renderCliOutputArgumentError(outputArguments));
+    writeStdout(renderCliOutputArgumentError(outputArguments));
     process.exitCode = 1;
   } else {
     let wroteOutput = false;
-    const resultOutput = createCliResultOutput(args, process.stdout);
-    await createCli(process.env, resultOutput).serve(args, {
+    const resultOutput = createCliResultOutput(cliArgs, process.stdout);
+    await createCli(process.env, resultOutput).serve(cliArgs, {
       stdout: (output) => {
         if (resultOutput?.handled) {
-          if (resultOutput.failed) process.stderr.write(output);
+          if (resultOutput.failed && !stdoutClosed)
+            process.stderr.write(output);
           return;
         }
         if (output.length > 0) wroteOutput = true;
-        process.stdout.write(output);
+        writeStdout(output);
       },
+      // A command failure caused by the closed pipe keeps the pipe's status.
+      exit: (code) => process.exit(stdoutClosed ? SIGPIPE_EXIT_STATUS : code),
     });
     if (!wroteOutput && !resultOutput?.handled)
-      process.stdout.write(renderEmptyFilteredCliOutput(args) ?? "");
+      writeStdout(renderEmptyFilteredCliOutput(cliArgs) ?? "");
   }
 }
 
+/**
+ * A reader that closes the pipe early, as `rea ... | head` does, has declined
+ * the rest of the output. Stop writing without reporting a failure, and exit
+ * with the status a SIGPIPE-terminated writer reports. MCP stdio keeps its own
+ * client lifecycle.
+ */
+function guardClosedStdout() {
+  process.stdout.on("error", (cause) => {
+    if (cause?.code === "EPIPE") {
+      stdoutClosed = true;
+      process.exitCode = SIGPIPE_EXIT_STATUS;
+      return;
+    }
+    // Other failures stay unhandled unless a pending JSON write reports them.
+    if (process.stdout.listenerCount("error") === 1) throw cause;
+  });
+}
+
+function writeStdout(output) {
+  if (!stdoutClosed) process.stdout.write(output);
+}
+
 async function compiledRuntimeExists(paths) {
   for (const path of paths) {
     try {

scripts/verify-linux-setup-e2e.mjs

@@ -13,6 +13,7 @@ import { promisify } from "node:util";
 
 import { completeVerifierRun, createVerifierRun } from "./lib/verifier-run.mjs";
 import { assertCleanLinuxHopperRuntime } from "./lib/clean-linux-hopper-runtime.mjs";
+import { isUnsupportedHopperLaunchFailure } from "./verify-package-platform.mjs";
 
 const execFileAsync = promisify(execFile);
 const verifierRun = createVerifierRun();
@@ -89,16 +90,24 @@ try {
   await appendFile(modifiedHopper, new Uint8Array([0]));
   await chmod(modifiedHopper, 0o700);
   try {
-    await execFileAsync(command, ["analyze", "/usr/bin/true"], {
-      env: { ...process.env, HOPPER_LAUNCHER_PATH: modifiedHopper },
-      timeout: 30_000,
-    });
+    await execFileAsync(
+      command,
+      ["analyze", "/usr/bin/true", "--provider", "hopper", "--json"],
+      {
+        env: { ...process.env, HOPPER_LAUNCHER_PATH: modifiedHopper },
+        timeout: 30_000,
+      },
+    );
     throw new Error("modified Hopper build was not rejected");
   } catch (error) {
-    const output = `${error?.stdout ?? ""}\n${error?.stderr ?? ""}`;
-    if (!output.includes("unsupported_hopper_build"))
+    if (
+      error?.code !== 1 ||
+      typeof error.stdout !== "string" ||
+      !isUnsupportedHopperLaunchFailure(JSON.parse(error.stdout))
+    )
       throw new Error(
         "modified Hopper build did not return its stable failure code",
+        { cause: error },
       );
   }
 } finally {

scripts/verify-package-environment.mjs

@@ -1,5 +1,6 @@
 import { chmod, mkdir, symlink, writeFile } from "node:fs/promises";
 import { join } from "node:path";
+import { pathToFileURL } from "node:url";
 import { supportedClients } from "../dist/application/SupportedClients.js";
 
 /** Create an isolated prefix, fake PATH, client configs, and environment. */
@@ -121,6 +122,14 @@ export const packageHopperEnvironment = (root, platform = process.platform) => {
     };
   return {
     HOPPER_LAUNCHER_PATH: "/bin/sh",
+    // Feed availability must not turn this foreign-launcher rejection test into
+    // an unrelated networking failure. Successful vendor verification is real E2E coverage.
+    NODE_OPTIONS: [
+      process.env.NODE_OPTIONS,
+      `--import=${pathToFileURL(join(root, "tests/fixtures/hopperRejectionFeed.mjs")).href}`,
+    ]
+      .filter(Boolean)
+      .join(" "),
     HOPPER_LOADER_ARGS_JSON: JSON.stringify([
       "-c",
       'node_path=$1; shift; "$node_path" "$@"',

scripts/verify-package-platform.mjs

@@ -88,14 +88,29 @@ export async function verifyPackagePlatform({
   }
 }
 
-/** Require the declared unsupported build or an exact missing-Xvfb diagnostic, never arbitrary provider failure. */
+/** One owner for unsupported-launch evidence across package and real-Hopper verification. */
+export const isUnsupportedHopperLaunchFailure = (failure) => {
+  const details = failure?.details;
+  const unsupportedBuild =
+    details?.failure_code === "unsupported_hopper_build" &&
+    details.exit_code === 72;
+  const publisherRejection =
+    details?.stage === "launch" &&
+    details.launch_rejection?.boundary === "publisher_verification" &&
+    details.launch_rejection.reason === "unsupported_hopper_build" &&
+    details.exit_code === undefined &&
+    details.launcher === undefined;
+  return (
+    failure?.code === "provider_unavailable" &&
+    (publisherRejection || unsupportedBuild)
+  );
+};
+
+/** Require an exact pre-execution rejection, observed helper exit, or missing-Xvfb diagnostic. */
 export const assertLinuxPackageProviderFailure = (execution) => {
   const failure = json(execution.stdout);
   const details = failure?.details;
   const diagnostic = details?.diagnostics;
-  const unsupportedBuild =
-    details?.failure_code === "unsupported_hopper_build" &&
-    details.exit_code === 72;
   const missingDisplayDependency =
     details?.failure_code === "runtime_dependency_unavailable" &&
     details.exit_code === 79 &&
@@ -108,7 +123,7 @@ export const assertLinuxPackageProviderFailure = (execution) => {
   if (
     execution.status !== 1 ||
     failure?.code !== "provider_unavailable" ||
-    (!unsupportedBuild && !missingDisplayDependency)
+    (!isUnsupportedHopperLaunchFailure(failure) && !missingDisplayDependency)
   )
     throw new Error(
       `packaged Linux Hopper verification did not fail closed: ${JSON.stringify(failure)}`,

scripts/verify-real-ghidra-windows.mjs

@@ -27,7 +27,9 @@ import {
 } from "../dist/ghidra/GhidraSessionValues.js";
 import { parseBinaryTarget } from "../dist/application/BinaryTargetResolver.js";
 import { completeVerifierRun, createVerifierRun } from "./lib/verifier-run.mjs";
+import { resolveGhidraVerifierTimeouts } from "./lib/ghidra-verifier-timeouts.mjs";
 import { verifyGhidraCompactReferences } from "./lib/real-ghidra-compact-references.mjs";
+import { verifyGhidraBorlandSeh } from "./lib/real-ghidra-borland-seh.mjs";
 import { verifyGhidraFileOffsets } from "./lib/real-ghidra-file-offsets.mjs";
 import { verifyWindowsGhidraByteSearch } from "./lib/ghidra-windows-byte-search.mjs";
 
@@ -47,6 +49,7 @@ if (
 )
   throw new Error("Windows Ghidra must not advertise database mutation");
 const installDir = process.env.GHIDRA_INSTALL_DIR;
+const { clientOptions } = resolveGhidraVerifierTimeouts(process.env);
 if (installDir === undefined || !isAbsolute(installDir))
   throw new Error(
     "Set GHIDRA_INSTALL_DIR to the absolute root of Ghidra 12.1.4.",
@@ -116,6 +119,7 @@ const client = new GhidraClient({
   transport: "authenticated-loopback-tcp",
   providerVersion: SUPPORTED_GHIDRA_VERSION,
   profileDigest: profile.value.profile.digest,
+  ...clientOptions,
 });
 
 let coordinates;
@@ -270,6 +274,10 @@ try {
         functionOperation,
         architecture === "x86",
       );
+  const borlandSeh =
+    architecture === "x86" && !dllFixture
+      ? await verifyGhidraBorlandSeh(functionObservationOperation, true)
+      : null;
   report = {
     ok: true,
     provider: { id: "ghidra", version: SUPPORTED_GHIDRA_VERSION },
@@ -286,6 +294,7 @@ try {
     byte_search: byteSearch,
     cleanup: "complete",
     compact_references: compactReferences,
+    borland_seh: borlandSeh,
     native_controls: {
       job_object: true,
       protected_dacl: true,
@@ -335,6 +344,11 @@ async function inventory(operation, parameters) {
 }
 
 async function functionOperation(operation, parameters) {
+  const observation = await functionObservationOperation(operation, parameters);
+  return observation.value;
+}
+
+async function functionObservationOperation(operation, parameters) {
   const input = parseGhidraFunctionInput(operation, parameters);
   if (!input.ok) throw input.error;
   const called = await client.callTool(operation, input.value);
@@ -342,7 +356,7 @@ async function functionOperation(operation, parameters) {
   const result = parseGhidraFunctionResult(operation, called.value);
   if (!result.ok) throw result.error;
   observed.add(operation);
-  return result.value.value;
+  return result.value;
 }
 
 function assertSession(session, targetSha256, profileDigest) {

scripts/verify-real-hopper-fat.mjs

@@ -23,7 +23,7 @@ import {
   requireMcpOperationResult,
 } from "./lib/mcp-verifier-results.mjs";
 import { snapshotHopperRuntime } from "./lib/real-hopper-cleanup.mjs";
-import { HOPPER_TARGET_LEASE_DIRECTORY } from "../dist/hopper/HopperTargetLease.js";
+import { HOPPER_TARGET_LEASE_DIRECTORY } from "../dist/process/HopperLease.js";
 import { parseConfig } from "../dist/config/parseConfig.js";
 import {
   verifyHopperSourceBinding,

scripts/verify-real-hopper.mjs

@@ -12,7 +12,7 @@ import {
   requirePseudocode,
 } from "../dist/application/RealHopperAssertions.js";
 import { HOPPER_PROVIDER_IDENTITY } from "../dist/hopper/HopperProviderCapabilities.js";
-import { HOPPER_TARGET_LEASE_DIRECTORY } from "../dist/hopper/HopperTargetLease.js";
+import { HOPPER_TARGET_LEASE_DIRECTORY } from "../dist/process/HopperLease.js";
 import { REA_WORKFLOW_PROVIDER } from "../dist/application/InvestigationProviders.js";
 import { loadRealHopperFixtureTargets } from "./lib/real-hopper-fixture.mjs";
 import { verifyHopperStringObjects } from "./lib/real-hopper-search.mjs";

scripts/verify-windows-native-inspection.mjs

@@ -55,14 +55,29 @@ try {
     join(workspace, "dist", "windows", "WindowsNativeLoader.js"),
   ).href;
   const source = `import assert from 'node:assert/strict';
+import {createRequire} from 'node:module';
 import {existsSync,mkdirSync,readFileSync,readdirSync,renameSync,statSync,writeFileSync} from 'node:fs';
 import {join} from 'node:path';
 import {systemWindowsNativeAuthority} from ${JSON.stringify(loader)};
 const temp=process.env.TEMP;
-const first=systemWindowsNativeAuthority();
-assert.equal(first.available,false);
-assert.equal(first.retryable,true);
-assert.match(first.reason,/Delete owned object by handle failed.*Win32 5/s);
+const mode=process.argv[2];
+const creation=mode.startsWith('create-');
+const native=creation?createRequire(import.meta.url)(${JSON.stringify(join(workspace, "native", "windows", "build", "rea-windows-x64.node"))}):null;
+if(creation){
+ let failure;
+ try{native.call('runtime_create',[temp,'rea-native-created-'])}catch(error){failure=error}
+ assert.ok(failure instanceof Error,'runtime_create unexpectedly passed injected post-create validation failure');
+ assert.equal(failure.constraint,'Read private DACL by handle failed');
+ assert.equal(failure.win32Code,5);
+ assert.equal(failure.cleanupConstraint,'Delete owned object by handle failed');
+ assert.equal(failure.cleanupWin32Code,5);
+ assert.notEqual(failure.runtimeAllocation,'not-started');
+} else {
+ const first=systemWindowsNativeAuthority();
+ assert.equal(first.available,false);
+ assert.equal(first.retryable,true);
+ assert.match(first.reason,/Delete owned object by handle failed.*Win32 5/s);
+}
 const [name,...extra]=readdirSync(temp);assert.ok(name);assert.equal(extra.length,0);
 const path=join(temp,name);const identity=statSync(path,{bigint:true});
 // Delete sharing permits rename; ownership follows the retained NTFS handle.
@@ -73,26 +88,40 @@ assert.equal(moved.dev,identity.dev);assert.equal(moved.ino,identity.ino);
 mkdirSync(path);const replacement=statSync(path,{bigint:true});
 assert.notEqual(replacement.ino,identity.ino);
 const sentinel=join(path,'unrelated.txt');writeFileSync(sentinel,'unrelated replacement');
-if(process.argv[2]!=='teardown'){
+if(mode==='retry'){
  const second=systemWindowsNativeAuthority();assert.equal(second.available,true);
  assert.equal(existsSync(renamed),false);
  assert.deepEqual(readdirSync(temp),[name]);
  assert.equal(readFileSync(sentinel,'utf8'),'unrelated replacement');
  assert.equal(statSync(path,{bigint:true}).ino,replacement.ino);
  assert.equal(systemWindowsNativeAuthority(),second);
+} else if(mode==='create-retry') {
+ const second=native.call('runtime_create',[temp,'rea-native-created-retry-']);
+ assert.equal(existsSync(renamed),false);
+ assert.equal(readFileSync(sentinel,'utf8'),'unrelated replacement');
+ native.call('runtime_close',[second.handle]);
+ assert.equal(existsSync(second.path),false);
+ assert.equal(statSync(path,{bigint:true}).ino,replacement.ino);
 }
-console.log(JSON.stringify({mode:process.argv[2],firstFailed:true,retainedOwner:true,name,renamed,device:identity.dev.toString(),inode:identity.ino.toString(),replacementInode:replacement.ino.toString()}));
+console.log(JSON.stringify({mode,firstFailed:true,retainedOwner:true,name,renamed,device:identity.dev.toString(),inode:identity.ino.toString(),replacementInode:replacement.ino.toString()}));
 `;
   const fixture = join(workspace, "probe.mjs");
   await writeFile(fixture, source);
-  for (const mode of ["retry", "teardown"]) {
+  for (const mode of ["retry", "teardown", "create-retry", "create-teardown"]) {
     const temp = join(workspace, mode);
     await mkdir(temp);
     const child = spawnSync(process.execPath, [fixture, mode], {
       encoding: "utf8",
       timeout: 20_000,
       windowsHide: true,
-      env: { ...process.env, TEMP: temp, TMP: temp },
+      env: {
+        ...process.env,
+        TEMP: temp,
+        TMP: temp,
+        ...(mode.startsWith("create-")
+          ? { REA_NATIVE_CREATION_FAILURE: "1" }
+          : {}),
+      },
     });
     if (child.error !== undefined) throw child.error;
     assert.equal(child.status, 0, child.stderr);
@@ -119,7 +148,7 @@ console.log(JSON.stringify({mode:process.argv[2],firstFailed:true,retainedOwner:
     );
   }
   process.stdout.write(
-    `${JSON.stringify({ ok: true, controlledDispositionFailure: true, retainedOwner: true, renamedOwnerCleansChildren: true, unrelatedReplacementPreserved: true, retryRecovers: true, successfulAdmissionCached: true, environmentTeardownCleansProbe: true, issue1701HostCauseReproduced: false })}\n`,
+    `${JSON.stringify({ ok: true, inspectionPostCreateValidationFailure: true, runtimeCreatePostCreateValidationFailure: true, controlledRollbackDispositionFailure: true, retainedOwner: true, renamedOwnerCleansChildren: true, unrelatedReplacementPreserved: true, retryRecoversByInspectionAndCreation: true, successfulAdmissionCached: true, environmentTeardownCleansProbeAndCreation: true, issue1701HostCauseReproduced: false })}\n`,
   );
 } finally {
   await rm(workspace, { recursive: true, force: true });

scripts/verify-windows-native.mjs

@@ -22,6 +22,7 @@ import { dirname, join, parse, resolve } from "node:path";
 import { pathToFileURL } from "node:url";
 import { verifyWindowsProtocolInput } from "./lib/windows-protocol-input.mjs";
 import { verifyWindowsJobMemoryLimit } from "./lib/windows-job-memory-limit.mjs";
+import { verifyVolumeRootAppendLease } from "./lib/windows-volume-root-append-lease.mjs";
 
 if (process.platform !== "win32")
   throw new Error(
@@ -47,6 +48,13 @@ const { PrivateRuntimeRootUnavailableError } = await import(
   )
 );
 const workspace = await mkdtemp(join(tmpdir(), "rea-native-conformance-"));
+// A pwsh caller can supply module paths incompatible with Windows PowerShell.
+// Each independent observer must discover its own module path.
+const windowsPowerShellEnvironment = Object.fromEntries(
+  Object.entries(process.env).filter(
+    ([name]) => name.toLowerCase() !== "psmodulepath",
+  ),
+);
 const environment = Object.entries(process.env)
   .filter(([, value]) => value !== undefined)
   .map(([key, value]) => `${key}=${value}`);
@@ -251,6 +259,17 @@ process.stdout.write(JSON.stringify({executable:basename(process.execPath),abiVe
   }
   const unrelatedFile = join(workspace, "unrelated-file");
   await writeFile(unrelatedFile, "preserve unrelated writes");
+  if (process.argv[3] !== undefined) {
+    await verifyVolumeRootAppendLease({
+      fixturePath: process.argv[3],
+      volumeRoot: parse(source).root,
+      verifyAdmission: () => {
+        const rootGuard = native.call("open", [source]);
+        native.call("close", [rootGuard.handle]);
+      },
+    });
+    report.controls.volumeRootAdmissionWithAppendHandle = true;
+  }
   const admitted = native.call("open", [source]);
   try {
     await assert.rejects(
@@ -297,6 +316,7 @@ process.stdout.write(JSON.stringify({executable:basename(process.execPath),abiVe
   ]);
   const sentinel = join(sourceDirectory, "sentinel");
   await writeFile(sentinel, "preserve");
+  const runtimeVerificationErrors = [];
   try {
     const acl = JSON.parse(
       execFileSync(
@@ -305,18 +325,13 @@ process.stdout.write(JSON.stringify({executable:basename(process.execPath),abiVe
           "-NoProfile",
           "-NonInteractive",
           "-Command",
-          "$a=Get-Acl -LiteralPath $env:REA_TEST_PRIVATE_RUNTIME_PATH; $u=[Security.Principal.WindowsIdentity]::GetCurrent().User; @{protected=$a.AreAccessRulesProtected; ownerMatchesUser=($a.GetOwner([Security.Principal.SecurityIdentifier]).Value -eq $u.Value); trusteesOnlyUserAndSystem=(@($a.GetAccessRules($true,$true,[Security.Principal.SecurityIdentifier]) | Where-Object {$_.IdentityReference.Value -ne $u.Value -and $_.IdentityReference.Value -ne 'S-1-5-18'}).Count -eq 0); count=@($a.Access).Count} | ConvertTo-Json -Compress",
+          "$ErrorActionPreference='Stop'; $p=$env:REA_TEST_PRIVATE_RUNTIME_PATH; $a=Get-Acl -LiteralPath $p; $u=[Security.Principal.WindowsIdentity]::GetCurrent().User; [IO.File]::SetAttributes($p,[IO.FileAttributes]([int][IO.File]::GetAttributes($p) -bor 1)); @{protected=$a.AreAccessRulesProtected; ownerMatchesUser=($a.GetOwner([Security.Principal.SecurityIdentifier]).Value -eq $u.Value); trusteesOnlyUserAndSystem=(@($a.GetAccessRules($true,$true,[Security.Principal.SecurityIdentifier]) | Where-Object {$_.IdentityReference.Value -ne $u.Value -and $_.IdentityReference.Value -ne 'S-1-5-18'}).Count -eq 0); count=@($a.Access).Count; rootReadonly=(([int][IO.File]::GetAttributes($p) -band 1) -ne 0)} | ConvertTo-Json -Compress",
         ],
         {
           encoding: "utf8",
-          // A pwsh caller can supply module paths incompatible with the
-          // Windows PowerShell observer. Let that observer build its own path.
+          timeout: 15_000,
           env: {
-            ...Object.fromEntries(
-              Object.entries(process.env).filter(
-                ([name]) => name.toLowerCase() !== "psmodulepath",
-              ),
-            ),
+            ...windowsPowerShellEnvironment,
             REA_TEST_PRIVATE_RUNTIME_PATH: runtime.path,
           },
         },
@@ -327,6 +342,7 @@ process.stdout.write(JSON.stringify({executable:basename(process.execPath),abiVe
       ownerMatchesUser: true,
       trusteesOnlyUserAndSystem: true,
       count: 2,
+      rootReadonly: true,
     });
     native.call("runtime_mkdir", [runtime.handle, "inputs"]);
     native.call("runtime_write", [
@@ -472,7 +488,10 @@ process.stdout.write(JSON.stringify({executable:basename(process.execPath),abiVe
         {
           encoding: "utf8",
           timeout: 15_000,
-          env: { ...process.env, REA_SNAPSHOT_READER_PATH: snapshotPath },
+          env: {
+            ...windowsPowerShellEnvironment,
+            REA_SNAPSHOT_READER_PATH: snapshotPath,
+          },
         },
       ).trim();
       assert.equal(readerDigest, readbackSnapshot.sha256);
@@ -562,12 +581,52 @@ process.stdout.write(JSON.stringify({executable:basename(process.execPath),abiVe
       asynchronousSnapshot: true,
       snapshotCancellation: true,
     };
-  } finally {
-    native.call("runtime_close", [runtime.handle]);
+    const readonlyChildren = JSON.parse(
+      execFileSync(
+        "powershell.exe",
+        [
+          "-NoProfile",
+          "-NonInteractive",
+          "-Command",
+          "$ErrorActionPreference='Stop'; $items=ConvertFrom-Json -InputObject $env:REA_TEST_READONLY_CHILDREN; @($items | ForEach-Object { $current=[int][IO.File]::GetAttributes($_); $next=if(($current -band 16) -ne 0){$current -bor 1}else{1}; [IO.File]::SetAttributes($_,[IO.FileAttributes]$next); $a=[int][IO.File]::GetAttributes($_); @{readonly=(($a -band 1) -ne 0); directory=(($a -band 16) -ne 0); onlyReadonly=($a -eq 1)} }) | ConvertTo-Json -Compress",
+        ],
+        {
+          encoding: "utf8",
+          timeout: 15_000,
+          env: {
+            ...windowsPowerShellEnvironment,
+            REA_TEST_READONLY_CHILDREN: JSON.stringify([
+              join(runtime.path, "inputs"),
+              join(runtime.path, "inputs", "target.bin"),
+            ]),
+          },
+        },
+      ),
+    );
+    assert.deepEqual(readonlyChildren, [
+      { readonly: true, directory: true, onlyReadonly: false },
+      { readonly: true, directory: false, onlyReadonly: true },
+    ]);
+  } catch (failure) {
+    runtimeVerificationErrors.push(failure);
   }
+  try {
+    native.call("runtime_close", [runtime.handle]);
+  } catch (cleanupFailure) {
+    runtimeVerificationErrors.push(cleanupFailure);
+  }
+  if (runtimeVerificationErrors.length === 1)
+    throw runtimeVerificationErrors[0];
+  if (runtimeVerificationErrors.length > 1)
+    throw new AggregateError(
+      runtimeVerificationErrors,
+      "Native runtime verification and cleanup failed",
+    );
   assert.equal(await exists(runtime.path), false);
   assert.equal(await readFile(sentinel, "utf8"), "preserve");
   report.controls.cleanupDoesNotTraverseJunctions = true;
+  report.controls.readonlyRootAndChildrenRemoved = true;
+  report.controls.readonlyFileWithoutArchiveAttributeRemoved = true;
   const fixture = join(workspace, "tree.mjs");
   await writeFile(
     fixture,

src/adb/AdbProvider.test.ts

@@ -709,7 +709,7 @@ it("projects package details from dumpsys package output", async () => {
   });
   expect(result).toMatchObject({
     version_name: "1.2.3",
-    version_code: 45,
+    version_code: "45",
     first_install_time: "2026-01-02 03:04:05",
     last_update_time: "2026-02-03 04:05:06",
     installer_package_name: "com.android.vending",
@@ -719,6 +719,30 @@ it("projects package details from dumpsys package output", async () => {
   });
 });
 
+// PackageInfo.getLongVersionCode() combines versionCodeMajor and versionCode
+// into a signed 64-bit long, which JSON numbers cannot hold exactly.
+it.each(["9007199254740993", "-4294967296"])(
+  "keeps the long version code %s exactly as dumpsys prints it",
+  async (versionCode) => {
+    const { binary } = await writeAdbStub(
+      shellStub({
+        "dumpsys:package": `Packages:
+  Package [com.example] (abcdef):
+    versionCode=${versionCode} minSdk=24 targetSdk=34
+`,
+      }),
+    );
+    const result = await evidenceResult(providerOver(binary), {
+      operation: "inspect_adb_package",
+      input: { serial: "emulator-5554", package: "com.example" },
+    });
+    expect(result).toMatchObject({
+      version_code: versionCode,
+      coverage: "complete",
+    });
+  },
+);
+
 // Android 10 and 11 append ": restricted=true" to hard- and soft-restricted
 // entries of the requested-permissions list (Settings.dumpPackageLPr).
 it("counts requested permissions that the device marks as restricted", async () => {

src/adb/AdbShellOutput.ts

@@ -193,7 +193,8 @@ export const parseWindowFocusOutput = (
 
 export interface AdbPackageDetails {
   readonly versionName: string | null;
-  readonly versionCode: number | null;
+  /** Signed 64-bit long version code exactly as Package Manager prints it. */
+  readonly versionCode: string | null;
   readonly firstInstallTime: string | null;
   readonly lastUpdateTime: string | null;
   readonly installerPackageName: string | null;
@@ -214,7 +215,7 @@ export const parseDumpsysPackageOutput = (
 ): AdbPackageDetails => {
   const details: {
     versionName: string | null;
-    versionCode: number | null;
+    versionCode: string | null;
     firstInstallTime: string | null;
     lastUpdateTime: string | null;
     installerPackageName: string | null;
@@ -252,8 +253,8 @@ export const parseDumpsysPackageOutput = (
     if (key === "versionName" && details.versionName === null)
       details.versionName = value === "" ? null : value;
     else if (key === "versionCode" && details.versionCode === null) {
-      const parsed = /^(\d+)/u.exec(value);
-      if (parsed !== null) details.versionCode = Number(parsed[1]!);
+      const parsed = /^(-?\d+)/u.exec(value);
+      if (parsed !== null) details.versionCode = parsed[1]!;
     } else if (key === "firstInstallTime" && details.firstInstallTime === null)
       details.firstInstallTime = value === "" ? null : value;
     else if (key === "lastUpdateTime" && details.lastUpdateTime === null)

src/apktool/ApktoolConfiguration.ts

@@ -8,10 +8,12 @@ import { snapshotEnvironment } from "../process/snapshotEnvironment.js";
 import { ABORTED, waitForAbortable } from "../process/AbortablePromise.js";
 import { ProviderStartupDeadline } from "../process/ProviderDeadline.js";
 
+import { randomUUID } from "node:crypto";
 import {
-  execFileOutput,
-  execFileOutputFailure,
-} from "../process/ExecFileOutput.js";
+  OwnedCommandFailure,
+  runOwnedCommand,
+} from "../process/OwnedCommand.js";
+import type { ProviderProcessSupervisor } from "../process/ProviderProcess.js";
 import type { ProviderAvailability } from "../application/AnalysisProvider.js";
 import type { ProviderRejectionCode } from "../contracts/providerSelection.js";
 import type { JsonValue } from "../domain/jsonValue.js";
@@ -152,14 +154,23 @@ export type ApktoolConfigurationFailureCode =
   | "version_unresolved"
   | "unsupported_version";
 
+/** Unresolved cleanup of a launched probe; its owner was retained separately. */
+export interface ApktoolProbeCleanup {
+  readonly reason: string;
+  readonly resources: readonly string[];
+}
+
 export class ApktoolConfigurationFailure extends Error {
+  readonly cleanup: ApktoolProbeCleanup | undefined;
+
   constructor(
     readonly code: ApktoolConfigurationFailureCode,
     message: string,
     readonly diagnostics: Readonly<Record<string, JsonValue>> = {},
-    options?: ErrorOptions,
+    options?: ErrorOptions & { readonly cleanup?: ApktoolProbeCleanup },
   ) {
     super(message, options);
+    this.cleanup = options?.cleanup;
   }
 }
 
@@ -228,6 +239,7 @@ export const inspectApktoolClient = async (
   signal?: AbortSignal,
   environment: Readonly<Record<string, string | undefined>> = process.env,
   platform: NodeJS.Platform = process.platform,
+  onCleanupOwner?: (owner: ProviderProcessSupervisor) => void,
 ): Promise<ApktoolClientIdentity> => {
   const hostEnvironment = snapshotEnvironment(environment, platform);
   let invocation: ApktoolInvocation;
@@ -254,23 +266,57 @@ export const inspectApktoolClient = async (
   let stdout: string;
   let stderr: string;
   try {
-    ({ stdout, stderr } = await execFileOutput(
-      invocation.command,
-      invocation.arguments,
+    // The owned process group stops a shell launcher's children with it; a
+    // child holding inherited output pipes cannot outlive the deadline.
+    const output = await runOwnedCommand(
       {
+        command: invocation.command,
+        arguments: invocation.arguments,
         windowsVerbatimArguments: invocation.windowsVerbatimArguments,
-        timeout: VERSION_TIMEOUT_MS,
-        stopSignal: "SIGTERM",
-        maxBuffer: VERSION_MAX_OUTPUT_BYTES,
-        env: hostEnvironment,
-        ...(signal === undefined ? {} : { signal }),
+        // Apktool launchers are commonly interpreter scripts, so ownership is
+        // proved by the run token rather than the launcher's argv prefix.
+        expectedCommand: null,
+        runId: randomUUID(),
+        hostEnvironment,
       },
-    ));
+      {
+        timeoutMs: VERSION_TIMEOUT_MS,
+        diagnosticBytes: VERSION_MAX_OUTPUT_BYTES,
+      },
+      signal === undefined ? {} : { signal },
+    );
+    stdout = output.stdout.text;
+    stderr = output.stderr.text;
   } catch (cause) {
-    const failure = execFileOutputFailure(cause);
-    if (signal?.aborted || failure?.code === "ABORT_ERR")
-      throw new AnalysisCancelledError("inspect_apktool_client", { cause });
-    if (failure?.code === "ENOENT")
+    if (
+      cause instanceof OwnedCommandFailure &&
+      cause.cleanupOwner !== undefined
+    )
+      onCleanupOwner?.(cause.cleanupOwner);
+    const cleanup =
+      cause instanceof OwnedCommandFailure && cause.cleanupFailure !== null
+        ? { reason: cause.cleanupFailure, resources: cause.resources }
+        : undefined;
+    const cleanupDetails = cleanup === undefined ? {} : { cleanup };
+    if (
+      signal?.aborted === true ||
+      (cause instanceof OwnedCommandFailure && cause.reason === "cancelled")
+    )
+      throw new AnalysisCancelledError("inspect_apktool_client", {
+        cause,
+        ...cleanupDetails,
+      });
+    if (cause instanceof OwnedCommandFailure && cause.reason === "timeout")
+      throw new AnalysisTimeoutError(
+        "inspect_apktool_client",
+        VERSION_TIMEOUT_MS,
+        { cause, ...cleanupDetails },
+      );
+    if (
+      cause instanceof Error &&
+      "code" in cause &&
+      (cause.code === "ENOENT" || cause.code === "EACCES")
+    )
       throw new ApktoolConfigurationFailure(
         "command_missing",
         `The selected apktool launcher was not found: ${selection.command}`,
@@ -281,19 +327,25 @@ export const inspectApktoolClient = async (
         },
         { cause },
       );
+    const snapshot =
+      cause instanceof OwnedCommandFailure ? cause.snapshot : null;
     throw new ApktoolConfigurationFailure(
       "version_unresolved",
       "The selected apktool launcher did not report a usable version",
       {
         selected_command: selection.command,
-        ...(failure === undefined
+        ...(snapshot === null
           ? {}
           : {
-              exit_code: failure.code,
-              stderr_excerpt: failure.stderr.slice(0, 512),
+              exit_code: snapshot.exitCode ?? null,
+              stderr_excerpt: snapshot.stderr.text.slice(0, 512),
             }),
+        ...(cause instanceof OwnedCommandFailure &&
+        cause.reason === "output-limit"
+          ? { max_output_bytes: VERSION_MAX_OUTPUT_BYTES }
+          : {}),
       },
-      { cause },
+      { cause, ...cleanupDetails },
     );
   }
   const identity = parseApktoolVersionOutput(stdout);

src/apktool/ApktoolProvider.ts

@@ -56,6 +56,7 @@ import { WindowsBatchCommandError } from "../process/WindowsBatchCommand.js";
 import {
   APKTOOL_COMMAND_ENV,
   ApktoolConfigurationFailure,
+  type ApktoolProbeCleanup,
   apktoolConfigurationAvailability,
   apktoolInvocation,
   inspectApktoolClient,
@@ -148,6 +149,19 @@ interface DecodeLimits {
   readonly maxOutputBytes: number;
 }
 
+/** A probe whose launched processes could not be confirmed stopped. */
+const probeCleanupFailure = (
+  operation: string,
+  failure: ApktoolConfigurationFailure,
+  cleanup: ApktoolProbeCleanup,
+): ProviderCleanupError =>
+  new ProviderCleanupError(
+    APKTOOL_PROVIDER_IDENTITY.id,
+    cleanup.resources,
+    { reason: cleanup.reason, probe_failure: failure.message },
+    { operation, cause: failure },
+  );
+
 const runApktool = async (
   operation: string,
   selection: ApktoolCommandSelection,
@@ -180,6 +194,9 @@ const runApktool = async (
         command: invocation.command,
         arguments: invocation.arguments,
         windowsVerbatimArguments: invocation.windowsVerbatimArguments,
+        // Apktool launchers are commonly interpreter scripts, so ownership is
+        // proved by the run token rather than the launcher's argv prefix.
+        expectedCommand: null,
         runId: randomUUID(),
         hostEnvironment: environment,
       },
@@ -328,6 +345,10 @@ export class ApktoolProvider {
   private readonly workspaces = new Set<string>();
   private readonly removeWorkspace: (path: string) => Promise<void>;
   readonly #pendingCommandOwners = new Map<ProviderProcessSupervisor, string>();
+  /** A version probe whose cleanup is unresolved is retried on close. */
+  readonly #retainProbeOwner = (owner: ProviderProcessSupervisor): void => {
+    this.#pendingCommandOwners.set(owner, "apktool version probe");
+  };
   readonly #shutdown = new AbortController();
   readonly #active = new Set<Promise<void>>();
   #closing = false;
@@ -378,7 +399,13 @@ export class ApktoolProvider {
         process.platform,
         signal,
       );
-      await inspectApktoolClient(selection, signal, environment);
+      await inspectApktoolClient(
+        selection,
+        signal,
+        environment,
+        process.platform,
+        this.#retainProbeOwner,
+      );
       throwIfCancelled("inspect_apktool_client", signal);
       return {
         status: "available",
@@ -390,8 +417,15 @@ export class ApktoolProvider {
         },
       };
     } catch (cause) {
-      if (cause instanceof ApktoolConfigurationFailure)
+      if (cause instanceof ApktoolConfigurationFailure) {
+        if (cause.cleanup !== undefined)
+          throw probeCleanupFailure(
+            "inspect_apktool_client",
+            cause,
+            cause.cleanup,
+          );
         return apktoolConfigurationAvailability(cause);
+      }
       throw cause;
     }
   }
@@ -437,7 +471,13 @@ export class ApktoolProvider {
         process.platform,
         signal,
       );
-      identity = await inspectApktoolClient(selection, signal, environment);
+      identity = await inspectApktoolClient(
+        selection,
+        signal,
+        environment,
+        process.platform,
+        this.#retainProbeOwner,
+      );
     } catch (cause) {
       if (
         cause instanceof AnalysisCancelledError ||
@@ -445,6 +485,10 @@ export class ApktoolProvider {
       )
         return err(cause);
       if (cause instanceof ApktoolConfigurationFailure) {
+        if (cause.cleanup !== undefined)
+          return err(
+            probeCleanupFailure(request.operation, cause, cause.cleanup),
+          );
         return err(
           providerConfigurationRefusal(
             APKTOOL_PROVIDER_IDENTITY.id,
@@ -497,12 +541,14 @@ export class ApktoolProvider {
   async #drainAndClean(): Promise<void> {
     await Promise.all(this.#active);
     const failures: unknown[] = [];
+    const retained: string[] = [];
     for (const [owner, root] of this.#pendingCommandOwners) {
       try {
         const stopped = await owner.stop();
         if (stopped.status === "incomplete") throw new Error(stopped.reason);
         this.#pendingCommandOwners.delete(owner);
       } catch (cause) {
+        retained.push(root);
         failures.push(
           new Error(
             `apktool process cleanup remains incomplete for ${root}: ${cause instanceof Error ? cause.message : String(cause)}`,
@@ -511,8 +557,23 @@ export class ApktoolProvider {
         );
       }
     }
+    // Ownership stays retained for a later close; report it as typed cleanup.
     if (failures.length > 0)
-      throw new AggregateError(failures, "Apktool process cleanup failed");
+      throw new ProviderCleanupError(
+        APKTOOL_PROVIDER_IDENTITY.id,
+        retained,
+        {
+          reason: failures
+            .map((failure) =>
+              failure instanceof Error ? failure.message : String(failure),
+            )
+            .join("; "),
+        },
+        {
+          operation: "close",
+          cause: new AggregateError(failures, "Apktool process cleanup failed"),
+        },
+      );
     for (const path of this.workspaces) {
       try {
         await this.removeWorkspace(path);

src/apktool/ApktoolVersionProbe.lifecycle.test.ts

@@ -0,0 +1,107 @@
+import { execFile } from "node:child_process";
+import { chmod, readFile, writeFile } from "node:fs/promises";
+import { join } from "node:path";
+import { setTimeout as delay } from "node:timers/promises";
+import { promisify } from "node:util";
+import { expect, it } from "vitest";
+
+import { AnalysisCancelledError } from "../domain/analysisErrorCore.js";
+import { createTestTempDirectory } from "../../tests/fixtures/temporaryDirectory.js";
+import { inspectApktoolClient } from "./ApktoolConfiguration.js";
+
+/**
+ * Only an exited process counts as stopped: it is gone, a zombie waiting to
+ * be reaped, or absent from `ps`. A failed inspection is not proof of exit.
+ */
+const alive = async (pid: number): Promise<boolean> => {
+  try {
+    process.kill(pid, 0);
+  } catch (cause: unknown) {
+    if ((cause as NodeJS.ErrnoException).code === "ESRCH") return false;
+    return true;
+  }
+  return promisify(execFile)("ps", ["-o", "stat=", "-p", `${pid}`]).then(
+    ({ stdout }) => {
+      const state = stdout.trim();
+      return state !== "" && !state.startsWith("Z");
+    },
+    // `ps` exits 1 when no process matches; any other failure is unknown.
+    (cause: unknown) => (cause as { code?: unknown }).code !== 1,
+  );
+};
+
+const waitFor = async (check: () => Promise<boolean>, ms: number) => {
+  const deadline = Date.now() + ms;
+  while (Date.now() < deadline) {
+    if (await check()) return true;
+    await delay(25);
+  }
+  return false;
+};
+
+const readPids = async (path: string): Promise<readonly number[]> =>
+  readFile(path, "utf8")
+    .then((text) => JSON.parse(text) as number[])
+    .catch(() => []);
+
+// The interpreter is the test runtime itself, so its environment (and the
+// ownership run token) is readable on both Linux and macOS.
+it.skipIf(process.platform === "win32")(
+  "stops an interpreter launcher and its pipe-holding child when the probe is cancelled",
+  async () => {
+    const root = await createTestTempDirectory("rea-apktool-probe-");
+    const pids = join(root, "pids.json");
+    const launcher = join(root, "apktool");
+    await writeFile(
+      launcher,
+      [
+        `#!${process.execPath}`,
+        'const { spawn } = require("node:child_process");',
+        'const child = spawn(process.execPath, ["-e", "setInterval(() => {}, 1000)"], { stdio: "inherit" });',
+        `require("node:fs").writeFileSync(${JSON.stringify(pids)}, JSON.stringify([process.pid, child.pid]));`,
+        "setInterval(() => {}, 1000);",
+      ].join("\n"),
+    );
+    await chmod(launcher, 0o755);
+    const controller = new AbortController();
+    const started = performance.now();
+    const probe = inspectApktoolClient(
+      { command: launcher, commandSource: "environment" },
+      controller.signal,
+      process.env,
+    ).catch((cause: unknown) => cause);
+    try {
+      expect(
+        await waitFor(async () => (await readPids(pids)).length === 2, 10_000),
+      ).toBe(true);
+      const launched = await readPids(pids);
+      controller.abort();
+      // A regressed probe never settles; bound the wait so cleanup still runs.
+      const outcome = await Promise.race([
+        probe,
+        delay(15_000).then(() => "still pending"),
+      ]);
+      expect(outcome).toBeInstanceOf(AnalysisCancelledError);
+      expect(performance.now() - started).toBeLessThan(15_000);
+      expect(
+        await waitFor(
+          async () =>
+            (await Promise.all(launched.map(alive))).every((live) => !live),
+          5_000,
+        ),
+      ).toBe(true);
+    } finally {
+      // A failed assertion must not leave the fixture's processes running.
+      controller.abort();
+      await Promise.race([probe, delay(10_000)]);
+      for (const pid of await readPids(pids))
+        if (await alive(pid))
+          try {
+            process.kill(pid, "SIGKILL");
+          } catch {
+            // Already exited between the check and the signal.
+          }
+    }
+  },
+  30_000,
+);

src/apktool/ApktoolVersionProbe.test.ts

@@ -0,0 +1,147 @@
+import { afterEach, expect, it, vi } from "vitest";
+
+import {
+  AnalysisCancelledError,
+  AnalysisTimeoutError,
+} from "../domain/analysisErrorCore.js";
+import { projectAnalysisError } from "../domain/analysisErrorProjection.js";
+import {
+  OwnedCommandFailure,
+  runOwnedCommand,
+} from "../process/OwnedCommand.js";
+import type { ProviderProcessSupervisor } from "../process/ProviderProcess.js";
+import { ProviderCleanupError } from "../domain/providerCleanupError.js";
+import {
+  ApktoolConfigurationFailure,
+  inspectApktoolClient,
+} from "./ApktoolConfiguration.js";
+import { ApktoolProvider } from "./ApktoolProvider.js";
+
+vi.mock("../process/OwnedCommand.js", async (importOriginal) => ({
+  ...(await importOriginal<typeof import("../process/OwnedCommand.js")>()),
+  runOwnedCommand: vi.fn(),
+}));
+afterEach(() => vi.mocked(runOwnedCommand).mockReset());
+
+const selection = {
+  command: "/opt/apktool/apktool",
+  commandSource: "environment",
+} as const;
+const owner = {} as ProviderProcessSupervisor;
+
+it("runs the version probe as an owned interpreter-script process group", async () => {
+  vi.mocked(runOwnedCommand).mockResolvedValueOnce({
+    stdout: { text: "2.9.3\n" },
+    stderr: { text: "" },
+  } as Awaited<ReturnType<typeof runOwnedCommand>>);
+  await expect(
+    inspectApktoolClient(selection, undefined, {}, "linux"),
+  ).resolves.toEqual({ version: "2.9.3" });
+  expect(vi.mocked(runOwnedCommand).mock.lastCall?.[0]).toMatchObject({
+    command: "/opt/apktool/apktool",
+    arguments: ["--version"],
+    expectedCommand: null,
+  });
+  expect(vi.mocked(runOwnedCommand).mock.lastCall?.[1]).toEqual({
+    timeoutMs: 20_000,
+    diagnosticBytes: 64 * 1024,
+  });
+});
+
+it("reports an expired version probe as a typed timeout", async () => {
+  vi.mocked(runOwnedCommand).mockRejectedValueOnce(
+    new OwnedCommandFailure("timeout", "Command deadline elapsed."),
+  );
+  const failure = await inspectApktoolClient(
+    selection,
+    undefined,
+    {},
+    "linux",
+  ).catch((cause: unknown) => cause);
+  expect(failure).toBeInstanceOf(AnalysisTimeoutError);
+  expect(projectAnalysisError(failure as AnalysisTimeoutError)).toMatchObject({
+    code: "provider_timeout",
+    category: "timeout",
+    details: { timeout_ms: 20_000 },
+  });
+});
+
+it("retains an unresolved probe cleanup owner and reports the cleanup", async () => {
+  vi.mocked(runOwnedCommand).mockRejectedValueOnce(
+    new OwnedCommandFailure(
+      "timeout",
+      "Command deadline elapsed.",
+      null,
+      "cleanup incomplete",
+      undefined,
+      ["pid:1"],
+      owner,
+    ),
+  );
+  const retained: ProviderProcessSupervisor[] = [];
+  const failure = await inspectApktoolClient(
+    selection,
+    undefined,
+    {},
+    "linux",
+    (cleanupOwner) => retained.push(cleanupOwner),
+  ).catch((cause: unknown) => cause);
+  expect(failure).toBeInstanceOf(AnalysisTimeoutError);
+  expect(projectAnalysisError(failure as AnalysisTimeoutError)).toMatchObject({
+    code: "cleanup_incomplete",
+  });
+  expect(retained).toEqual([owner]);
+});
+
+it("reports a cancelled version probe as cancelled", async () => {
+  vi.mocked(runOwnedCommand).mockRejectedValueOnce(
+    new OwnedCommandFailure("cancelled", "Command cancelled."),
+  );
+  await expect(
+    inspectApktoolClient(selection, undefined, {}, "linux"),
+  ).rejects.toBeInstanceOf(AnalysisCancelledError);
+});
+
+it("keeps unresolved probe cleanup typed when the launcher overflows its output", async () => {
+  const overflow = () =>
+    new OwnedCommandFailure(
+      "output-limit",
+      "Command diagnostic output exceeded its complete-output budget.",
+      null,
+      "process group still running",
+      undefined,
+      ["pid:7", "process-group:7"],
+      owner,
+    );
+  vi.mocked(runOwnedCommand).mockRejectedValueOnce(overflow());
+  const failure = await inspectApktoolClient(
+    selection,
+    undefined,
+    {},
+    "linux",
+  ).catch((cause: unknown) => cause);
+  expect(failure).toBeInstanceOf(ApktoolConfigurationFailure);
+  expect((failure as ApktoolConfigurationFailure).cleanup).toEqual({
+    reason: "process group still running",
+    resources: ["pid:7", "process-group:7"],
+  });
+
+  // Both provider entry points report the unresolved processes as cleanup.
+  const provider = new ApktoolProvider({
+    environment: { REA_APKTOOL_COMMAND: "/opt/apktool/apktool" },
+  });
+  vi.mocked(runOwnedCommand).mockRejectedValueOnce(overflow());
+  await expect(provider.inspectAvailability()).rejects.toBeInstanceOf(
+    ProviderCleanupError,
+  );
+  vi.mocked(runOwnedCommand).mockRejectedValueOnce(overflow());
+  const result = await provider.execute({
+    operation: "decode_android_resources",
+    input: {},
+  } as never);
+  expect(result.ok ? null : result.error).toBeInstanceOf(ProviderCleanupError);
+  expect(result.ok ? null : projectAnalysisError(result.error)).toMatchObject({
+    code: "cleanup_incomplete",
+    details: { resources: ["pid:7", "process-group:7"] },
+  });
+});

src/application/CapabilityInventory.ts

@@ -22,6 +22,7 @@ import type { ToolKind } from "../contracts/toolContractTypes.js";
 import type { JsonValue } from "../domain/jsonValue.js";
 import { isMachOFormat } from "../domain/binaryTargetTypes.js";
 import type { ProviderAvailability } from "./AnalysisProvider.js";
+import type { JebAnalysisAvailability } from "./jeb/JebAnalysisPort.js";
 import { TOOL_CONTRACTS } from "../contracts/toolContracts.js";
 import { FRIDA_TOOL_CONTRACTS } from "../contracts/fridaToolContracts.js";
 import {
@@ -56,7 +57,7 @@ export type AvailabilityPolicy = {
   readonly androidAnalysisAvailability?: ProviderAvailability;
   readonly adbAnalysisAvailability?: ProviderAvailability;
   readonly apktoolAnalysisAvailability?: ProviderAvailability;
-  readonly jebAnalysisAvailability?: ProviderAvailability;
+  readonly jebAnalysisAvailability?: JebAnalysisAvailability;
   readonly javascriptRecoveryEnabled?: boolean;
   readonly webModuleResolutionEnabled?: boolean;
   readonly browserObservationEnabled?: boolean;
@@ -485,15 +486,22 @@ const workflowAvailabilityFor = ({
       policy.apktoolAnalysisAvailability,
       "Set REA_APKTOOL_COMMAND to a caller-supplied Apktool executable or make Apktool 2.7.0 or newer available on PATH.",
     );
-  if (kind === "jeb-provider")
-    return policy.jebAnalysisAvailability?.status === "available"
-      ? { reason: "available", remediation: null }
-      : {
-          reason: "provider_missing",
-          remediation:
-            policy.jebAnalysisAvailability?.reason ??
-            "Start a JEB client serving MCP (GUI client, or headless via its programmatic launcher) and point REA_JEB_MCP_URL at it (default http://127.0.0.1:8425/mcp). REA does not install or launch JEB.",
-        };
+  if (kind === "jeb-provider") {
+    const availability = policy.jebAnalysisAvailability;
+    if (availability?.status === "available")
+      return availability.supportedOperations.some(
+        (operation) => operation === name,
+      )
+        ? { reason: "available", remediation: null }
+        : {
+            reason: "provider_unavailable",
+            remediation: `The running JEB endpoint does not advertise support for ${name}. Use a JEB client build that provides this operation.`,
+          };
+    return ownerProbeAvailability(
+      availability,
+      "Start a JEB client serving MCP (GUI client, or headless via its programmatic launcher) and point REA_JEB_MCP_URL at it (default http://127.0.0.1:8425/mcp). REA does not install or launch JEB.",
+    );
+  }
   const browser = browserProviderAvailability(name, kind, policy);
   if (browser !== null) return browser;
   if (name === "capture_process_scenario" && !policy.processCaptureEnabled)

src/application/ClientRegistrationEnvironment.ts

@@ -13,6 +13,18 @@ import {
 import { PRODUCT_IDENTITY } from "../identity.js";
 import { compareUnicodeCodePoints } from "../domain/unicodeCodePointOrder.js";
 import { dshDocument } from "./DshCordisPatch.js";
+import type { McpToolTimeoutInputs } from "../mcpStartupPolicy.js";
+import type { AnalysisProviderSelector } from "../contracts/providerSelection.js";
+
+export type ProviderRequestBudgetResolver = (
+  analysisProvider: AnalysisProviderSelector,
+  environment: Readonly<Record<string, string>>,
+) => Result<number | undefined, ConfigurationError>;
+
+export interface ClientRegistrationEnvironmentDetails {
+  readonly values: Readonly<Record<string, string>>;
+  readonly timeoutInputs: McpToolTimeoutInputs;
+}
 
 /** Read explicit REA overrides without dropping peers when one value is invalid. */
 export const clientRegistrationEnvironment = (
@@ -22,6 +34,33 @@ export const clientRegistrationEnvironment = (
 ): Result<Readonly<Record<string, string>>, ConfigurationError> => {
   if (parsed?.dialect === "dsh")
     return dshDocument(parsed.document).environment(overrides, validation);
+  const details = clientRegistrationEnvironmentDetails(
+    parsed,
+    overrides,
+    validation,
+  );
+  return details.ok ? ok(details.value.values) : details;
+};
+
+/** Retain original strings and the parser-owned timeout facts. */
+export const clientRegistrationEnvironmentDetails = (
+  parsed: ClientConfigurationDocument | undefined,
+  overrides: Readonly<Record<string, string>> = {},
+  validation: "fragment" | "effective" = "effective",
+  resolveProviderRequestBudget?: ProviderRequestBudgetResolver,
+): Result<ClientRegistrationEnvironmentDetails, ConfigurationError> => {
+  if (parsed?.dialect === "dsh") {
+    const environment = dshDocument(parsed.document).environment(
+      overrides,
+      validation,
+    );
+    if (!environment.ok) return environment;
+    return validatedDetails(
+      environment.value,
+      validation,
+      resolveProviderRequestBudget,
+    );
+  }
   const key =
     parsed?.dialect === "opencode" || parsed?.dialect === "opencode_v2"
       ? "environment"
@@ -50,16 +89,45 @@ export const clientRegistrationEnvironment = (
   // Validate known settings with the runtime's parser, but retain the original
   // string overrides, including unknown client-specific keys and no defaults.
   const environment = { ...existing.data, ...overrides };
+  return validatedDetails(
+    environment,
+    validation,
+    resolveProviderRequestBudget,
+  );
+};
+
+const validatedDetails = (
+  environment: Readonly<Record<string, string>>,
+  validation: "fragment" | "effective",
+  resolveProviderRequestBudget?: ProviderRequestBudgetResolver,
+): Result<ClientRegistrationEnvironmentDetails, ConfigurationError> => {
   const validated =
     validation === "fragment"
       ? parseEnvironmentValues(environment)
       : parseEnvironment(environment);
   if (!validated.ok) return validated;
-  return ok(
-    Object.fromEntries(
+  const providerRequestBudget =
+    resolveProviderRequestBudget?.(
+      validated.value.REA_ANALYSIS_PROVIDER,
+      environment,
+    ) ?? ok(undefined);
+  if (!providerRequestBudget.ok) return providerRequestBudget;
+  return ok({
+    values: Object.fromEntries(
       Object.entries(environment).sort(([left], [right]) =>
         compareUnicodeCodePoints(left, right),
       ),
     ),
-  );
+    timeoutInputs: {
+      analysisProvider: validated.value.REA_ANALYSIS_PROVIDER,
+      ghidraStartupTimeoutMs: validated.value.REA_GHIDRA_STARTUP_TIMEOUT_MS,
+      hopperStartupTimeoutMs: validated.value.REA_HOPPER_STARTUP_TIMEOUT_MS,
+      ...(validated.value.REA_JEB_MCP_TIMEOUT_MS === undefined
+        ? {}
+        : { jebMcpTimeoutMs: validated.value.REA_JEB_MCP_TIMEOUT_MS }),
+      ...(providerRequestBudget.value === undefined
+        ? {}
+        : { providerRequestTimeoutMs: providerRequestBudget.value }),
+    },
+  });
 };

src/application/ClientRegistrationIdentity.ts

@@ -13,8 +13,9 @@ const environmentValue = (
   platform: NodeJS.Platform,
 ): string | undefined => {
   if (platform !== "win32") return environment[name];
-  const key = Object.keys(environment).find(
-    (candidate) => candidate.toLowerCase() === name,
+  const target = name.toLowerCase();
+  const key = Object.keys(environment).findLast(
+    (candidate) => candidate.toLowerCase() === target,
   );
   return key === undefined ? undefined : environment[key];
 };

src/application/ClientRegistrationStatus.ts

@@ -14,7 +14,13 @@ import { valid } from "semver";
 
 import { PRODUCT_IDENTITY } from "../identity.js";
 import { CURRENT_ENTRYPOINT_PATH } from "../process/CurrentEntrypointPath.js";
-import { MCP_STARTUP_POLICY } from "../mcpStartupPolicy.js";
+import {
+  MCP_STARTUP_POLICY,
+  mcpProviderDeadlineAllowanceMs,
+  recommendedMcpToolTimeoutMs,
+  recommendedMcpToolTimeoutSeconds,
+  type McpToolTimeoutInputs,
+} from "../mcpStartupPolicy.js";
 import {
   clientRegistrationPackageReference,
   classifyClientRegistrationCommand,
@@ -28,6 +34,8 @@ import {
 import type { SetupClient } from "./SupportedClients.js";
 import { compareUnicodeCodePoints } from "../domain/unicodeCodePointOrder.js";
 import { dshDocument } from "./DshCordisPatch.js";
+import { clientRegistrationEnvironmentDetails } from "./ClientRegistrationEnvironment.js";
+import type { ProviderRequestBudgetResolver } from "./ClientRegistrationEnvironment.js";
 import type { ArtifactResourceScope } from "../artifacts/ArtifactResourceScope.js";
 
 interface ClientRegistrationStatusBase {
@@ -35,6 +43,8 @@ interface ClientRegistrationStatusBase {
   readonly config_path: string;
   /** Exact package reference requested by npx/bunx, when registration uses one. */
   readonly package_reference?: string;
+  /** Non-blocking timeout guidance; caller-selected values remain aligned. */
+  readonly advisory?: string;
 }
 
 type RegistrationCommand = readonly [string, ...string[]];
@@ -73,6 +83,7 @@ const registrationSchema = z
     command: z.string().min(1),
     args: z.array(z.string()).default([]),
     startup_timeout_sec: z.number().positive().optional(),
+    tool_timeout_sec: z.number().positive().optional(),
     startupTimeoutMs: z.number().positive().optional(),
     toolTimeoutMs: z.number().positive().optional(),
     type: z.string().optional(),
@@ -88,6 +99,7 @@ interface ClientRegistrationReadOptions {
   readonly platform?: NodeJS.Platform;
   readonly environment?: NodeJS.ProcessEnv;
   readonly resources?: ArtifactResourceScope;
+  readonly resolveProviderRequestBudget?: ProviderRequestBudgetResolver;
 }
 
 /** Registration facts shared by diagnostics and update maintenance. */
@@ -215,6 +227,28 @@ export const readClientRegistrationInspections = async (
             }
           : client;
       const registration = parseRegistration(raw, effectiveClient);
+      const registrationEnvironment =
+        client.name === "kimi" || client.name === "codex"
+          ? clientRegistrationEnvironmentDetails(
+              parsed,
+              {},
+              "effective",
+              options.resolveProviderRequestBudget,
+            )
+          : undefined;
+      if (
+        registrationEnvironment !== undefined &&
+        !registrationEnvironment.ok
+      ) {
+        record({
+          client: client.name,
+          config_path: configPath,
+          command: [],
+          state: "invalid",
+          remediation: registrationEnvironment.error.message,
+        });
+        continue;
+      }
       const command: RegistrationCommand = [
         registration.command,
         ...registration.args,
@@ -225,12 +259,16 @@ export const readClientRegistrationInspections = async (
       const packageVersionDrift =
         packageVersion !== undefined &&
         packageVersion !== PRODUCT_IDENTITY.packageVersion;
-      const alignment = registrationAligned(
+      const alignment = registrationAligned({
         registration,
-        effectiveClient,
+        client: effectiveClient,
         currentCommandPath,
-        options.platform ?? process.platform,
-      );
+        platform: options.platform ?? process.platform,
+        kimiInputSchemaProfileSelected:
+          registrationEnvironment?.ok === true &&
+          registrationEnvironment.value.values.REA_MCP_INPUT_SCHEMA_PROFILE !==
+            undefined,
+      });
       record(
         configuredStatus(
           effectiveClient,
@@ -252,6 +290,13 @@ export const readClientRegistrationInspections = async (
                 ? unknownEntryPointRemediation()
                 : undefined),
           packageReference,
+          registrationEnvironment?.ok === true
+            ? timeoutAdvisory(
+                client,
+                registration,
+                registrationEnvironment.value.timeoutInputs,
+              )
+            : undefined,
         ),
         activation,
       );
@@ -284,12 +329,19 @@ const registrationActivation = (
     ? "disabled"
     : "enabled";
 
-const registrationAligned = (
-  registration: z.output<typeof registrationSchema>,
-  client: SetupClient,
-  currentCommandPath: string | undefined,
-  platform: NodeJS.Platform,
-): "aligned" | "stale" | "unknown" => {
+const registrationAligned = ({
+  registration,
+  client,
+  currentCommandPath,
+  platform,
+  kimiInputSchemaProfileSelected,
+}: {
+  readonly registration: z.output<typeof registrationSchema>;
+  readonly client: SetupClient;
+  readonly currentCommandPath: string | undefined;
+  readonly platform: NodeJS.Platform;
+  readonly kimiInputSchemaProfileSelected: boolean;
+}): "aligned" | "stale" | "unknown" => {
   const command = [registration.command, ...registration.args];
   const identity = classifyClientRegistrationCommand(
     command,
@@ -298,18 +350,24 @@ const registrationAligned = (
   if (identity === "unknown") return "unknown";
   if (identity === "foreign") return "stale";
   if (
-    (client.name === "codex" || client.name === "grok_build") &&
+    client.name === "grok_build" &&
     registration.startup_timeout_sec !==
       MCP_STARTUP_POLICY.codexStartupTimeoutSeconds
   )
     return "stale";
   if (
-    client.name === "kimi" &&
-    (registration.startupTimeoutMs !==
-      MCP_STARTUP_POLICY.kimiStartupTimeoutMs ||
-      registration.toolTimeoutMs !== MCP_STARTUP_POLICY.kimiToolTimeoutMs)
+    client.name === "codex" &&
+    (registration.startup_timeout_sec === undefined ||
+      registration.tool_timeout_sec === undefined)
   )
     return "stale";
+  if (
+    client.name === "kimi" &&
+    (registration.startupTimeoutMs === undefined ||
+      registration.toolTimeoutMs === undefined)
+  )
+    return "stale";
+  if (client.name === "kimi" && !kimiInputSchemaProfileSelected) return "stale";
   if (
     (client.format === "kimi" ||
       client.format === "omp" ||
@@ -435,6 +493,7 @@ const configuredStatus = (
   state: "aligned" | "stale" | "unknown",
   policyRemediation?: string,
   packageReference?: string,
+  advisory?: string,
 ): ClientRegistrationStatus =>
   state === "aligned"
     ? {
@@ -446,6 +505,7 @@ const configuredStatus = (
           : { package_reference: packageReference }),
         state,
         remediation: null,
+        ...(advisory === undefined ? {} : { advisory }),
       }
     : {
         client: client.name,
@@ -458,6 +518,35 @@ const configuredStatus = (
         remediation: policyRemediation ?? remediation(client),
       };
 
+const timeoutAdvisory = (
+  client: SetupClient,
+  registration: z.output<typeof registrationSchema>,
+  timeoutInputs: McpToolTimeoutInputs,
+): string | undefined => {
+  const toolTimeout =
+    client.name === "kimi"
+      ? registration.toolTimeoutMs
+      : client.name === "codex"
+        ? registration.tool_timeout_sec
+        : undefined;
+  if (toolTimeout === undefined) return undefined;
+  const providerAllowance = mcpProviderDeadlineAllowanceMs(timeoutInputs);
+  const isKimi = client.name === "kimi";
+  const toolTimeoutMs = isKimi ? toolTimeout : toolTimeout * 1000;
+  if (toolTimeoutMs >= providerAllowance) return undefined;
+  const setting = isKimi ? "toolTimeoutMs" : "tool_timeout_sec";
+  const currentValue = isKimi
+    ? `${String(toolTimeout)} ms`
+    : `${String(toolTimeout)} seconds`;
+  const recommendedValue = isKimi
+    ? `${String(recommendedMcpToolTimeoutMs(timeoutInputs))} ms`
+    : `${String(recommendedMcpToolTimeoutSeconds(timeoutInputs))} seconds`;
+  const headroom = isKimi
+    ? `${String(MCP_STARTUP_POLICY.toolOperationHeadroomMs)} ms`
+    : `${String(MCP_STARTUP_POLICY.toolOperationHeadroomMs / 1000)} seconds`;
+  return `The configured ${setting} (${currentValue}) is shorter than REA's known provider allowance (${String(providerAllowance)} ms), so provider startup or an upstream request may exceed this client deadline. REA recommends ${setting} = ${recommendedValue} to cover that allowance plus ${headroom} of operation headroom; this additional headroom is a recommendation, not a guarantee for every analysis.`;
+};
+
 const exactRegisteredPackageVersion = (
   packageReference: string | undefined,
 ): string | undefined => {

src/application/Doctor.fixture.ts

@@ -1,5 +1,6 @@
 import { CATALOG_IDENTITY } from "../catalogIdentity.js";
 import { PRODUCT_IDENTITY } from "../identity.js";
+import { ok } from "../domain/result.js";
 import type { DoctorHost } from "./Doctor.js";
 
 /** Healthy recording host for focused doctor and projection tests. */
@@ -14,7 +15,16 @@ export const createDoctorHostFixture = (
   linuxDistribution: () => Promise.resolve(undefined),
   validTarget: (path) => Promise.resolve(path.includes("Hopper")),
   executable: (path) => Promise.resolve(path.includes("Hopper")),
-  supportedLinuxHopper: () => Promise.resolve(true),
+  inspectLinuxHopper: () =>
+    Promise.resolve(
+      ok({
+        sha256: "0".repeat(64),
+        version: "fixture",
+        primaryFingerprint: "0".repeat(40),
+        signingFingerprint: "1".repeat(40),
+        source: "signed_package_cache" as const,
+      }),
+    ),
   linuxDemoRuntimeCheck: () =>
     Promise.resolve({
       name: "hopper-demo-runtime",

src/application/Doctor.test.ts

@@ -4,6 +4,8 @@ import { SUPPORTED_NODE_VERSION_PROSE } from "../domain/runtimeVersion.js";
 import type { ClientRegistrationStatus } from "./ClientRegistrationStatus.js";
 import { createDoctorHostFixture as host } from "./Doctor.fixture.js";
 import { runDoctor } from "./Doctor.js";
+import { err } from "../domain/result.js";
+import { HopperReleaseError } from "./HopperReleaseFeed.js";
 
 describe("doctor runtime check", () => {
   it("reports Node 25 as unsupported", async () => {
@@ -363,6 +365,38 @@ describe("doctor scoped readiness", () => {
     );
     expect(staleSelected.healthy).toBe(false);
   });
+
+  it("keeps a caller-selected short timeout healthy and reports its advisory as informational", async () => {
+    const result = await runDoctor(
+      undefined,
+      host({
+        clientRegistrations: () =>
+          Promise.resolve([
+            {
+              client: "kimi",
+              config_path: "/home/user/.kimi-code/mcp.json",
+              command: ["/current/rea", "mcp"],
+              state: "aligned",
+              remediation: null,
+              advisory: "The configured toolTimeoutMs is shorter than startup.",
+            },
+          ]),
+      }),
+      { clients: ["kimi"] },
+    );
+
+    expect(result.healthy).toBe(true);
+    expect(result.scope_checks).toContainEqual(
+      expect.objectContaining({ name: "registration:kimi", ok: true }),
+    );
+    expect(result.informational_checks).toContainEqual(
+      expect.objectContaining({
+        name: "registration_timeout:kimi",
+        ok: true,
+        detail: "The configured toolTimeoutMs is shorter than startup.",
+      }),
+    );
+  });
 });
 
 describe("Grok Bot readiness", () => {
@@ -585,12 +619,20 @@ describe("doctor Linux and Hopper discovery", () => {
             hopperSetupSupported: true,
           }),
         executable: (candidate) => Promise.resolve(candidate === path),
-        supportedLinuxHopper: () => Promise.resolve(false),
+        inspectLinuxHopper: () =>
+          Promise.resolve(
+            err(
+              new HopperReleaseError(
+                "unsupported_hopper_build",
+                "Unverified fixture build",
+              ),
+            ),
+          ),
       }),
     );
     expect(
       result.checks.find(({ name }) => name === "hopper-version")?.remediation,
-    ).toContain("Unset or update HOPPER_LAUNCHER_PATH");
+    ).toContain("Check HOPPER_LAUNCHER_PATH");
   });
 });
 

src/application/Doctor.ts

@@ -19,11 +19,14 @@ import {
   macHopperRenamedLauncherCandidates,
 } from "./MacHopperPaths.js";
 import {
-  linuxHopperBinarySupported,
+  inspectLinuxHopperLauncher,
   linuxSharedLibrariesAvailable,
   readLinuxDistribution,
   type LinuxDistribution,
 } from "./LinuxHopper.js";
+import type { Result } from "../domain/result.js";
+import type { HopperReleaseError } from "./HopperReleaseFeed.js";
+import type { VerifiedLinuxHopperLauncher } from "./LinuxHopperPublisher.js";
 import {
   readInstalledSkillIdentity,
   type InstalledSkillIdentity,
@@ -36,6 +39,7 @@ import {
   type ClientRegistrationStatus,
   type UnhealthyClientRegistrationStatus,
 } from "./ClientRegistrationStatus.js";
+import type { ProviderRequestBudgetResolver } from "./ClientRegistrationEnvironment.js";
 import {
   collectDoctorDiagnostics,
   doctorHealthy,
@@ -100,11 +104,15 @@ export interface DoctorHost {
   /** Route-specific target admission details, when supplied by the host. */
   inspectTarget?(path: string): Promise<DoctorCheck>;
   executable(path: string): Promise<boolean>;
-  supportedLinuxHopper(path: string): Promise<boolean>;
+  inspectLinuxHopper(
+    path: string,
+  ): Promise<Result<VerifiedLinuxHopperLauncher, HopperReleaseError>>;
   linuxDemoRuntimeCheck(): Promise<DoctorCheck>;
   brewHopperPath(): Promise<string | undefined>;
   manualHopperPaths(): Promise<readonly string[]>;
   providerInspections?(): Promise<readonly DoctorProviderInspection[]>;
+  /** Resolve a parsed upstream request budget from the effective client environment. */
+  resolveProviderRequestBudget?: ProviderRequestBudgetResolver;
   /** Validated provider settings preserved by setup, independent of readiness. */
   providerRegistrationEnvironment?(): Readonly<Record<string, string>>;
   installationPaths?(): Promise<readonly string[]>;
@@ -220,7 +228,11 @@ const inspectDoctor = async (
     providerInspections,
   } = await collectDoctorDiagnostics(target, host);
   const identity = await collectDoctorIdentity(host, scope);
-  const checks = [...diagnosticChecks, ...identity.checks];
+  const checks = [
+    ...diagnosticChecks,
+    ...identity.checks,
+    ...registrationAdvisoryChecks(identity.value.registrations),
+  ];
   const providers = {
     hopperAvailable:
       hopperPath !== undefined &&
@@ -252,6 +264,22 @@ const inspectDoctor = async (
   };
 };
 
+const registrationAdvisoryChecks = (
+  registrations: readonly ClientRegistrationStatus[],
+): readonly DoctorCheck[] =>
+  registrations.flatMap(({ client, advisory }) =>
+    advisory === undefined
+      ? []
+      : [
+          {
+            name: `registration_timeout:${client}`,
+            ok: true,
+            classification: "healthy" as const,
+            detail: advisory,
+          },
+        ],
+  );
+
 const closeDoctorHost = async (
   host: DoctorHost,
 ): Promise<string | undefined> => {
@@ -411,6 +439,7 @@ export interface SystemDoctorHostOptions {
   readonly providerRegistrationEnvironment?: () => Readonly<
     Record<string, string>
   >;
+  readonly resolveProviderRequestBudget?: ProviderRequestBudgetResolver;
   readonly linuxDemoRuntimeCheck?: () => Promise<DoctorCheck>;
 }
 
@@ -488,7 +517,7 @@ export const systemDoctorHost = (
       ),
     linuxDemoRuntimeCheck:
       options.linuxDemoRuntimeCheck ?? uncomposedLinuxDemoRuntimeCheck,
-    supportedLinuxHopper: linuxHopperBinarySupported,
+    inspectLinuxHopper: (path) => inspectLinuxHopperLauncher(path, environment),
     async brewHopperPath() {
       return probeHomebrew(async (command) => {
         let caskroom: string;
@@ -527,6 +556,9 @@ export const systemDoctorHost = (
           providerRegistrationEnvironment:
             options.providerRegistrationEnvironment,
         }),
+    ...(options.resolveProviderRequestBudget === undefined
+      ? {}
+      : { resolveProviderRequestBudget: options.resolveProviderRequestBudget }),
     async ilspyCmdVersion(path) {
       try {
         await access(path, constants.X_OK);
@@ -579,6 +611,12 @@ export const systemDoctorHost = (
         platform,
         environment,
         resources,
+        ...(options.resolveProviderRequestBudget === undefined
+          ? {}
+          : {
+              resolveProviderRequestBudget:
+                options.resolveProviderRequestBudget,
+            }),
       }),
   };
 };

src/application/DoctorDiagnostics.ts

@@ -209,17 +209,34 @@ const optionalHopperVersionCheck = async (
   hopperPath: string | undefined,
 ): Promise<DoctorCheck | undefined> => {
   if (host.platform !== "linux" || hopperPath === undefined) return undefined;
-  return check(
-    "hopper-version",
-    await host.supportedLinuxHopper(hopperPath),
-    hopperPath,
-    {
-      remediation: unsupportedHopperRemediation(
-        host.configuredHopperPath === hopperPath,
-      ),
-      classification: "config_drift",
-    },
-  );
+  const inspected = await host.inspectLinuxHopper(hopperPath);
+  const buildCheck = check("hopper-version", inspected.ok, hopperPath, {
+    remediation: unsupportedHopperRemediation(
+      host.configuredHopperPath === hopperPath,
+    ),
+    classification:
+      !inspected.ok &&
+      [
+        "release_metadata",
+        "download",
+        "verification_unavailable",
+        "cancelled",
+        "cleanup_failed",
+      ].includes(inspected.error.reason)
+        ? "unknown"
+        : "config_drift",
+  });
+  return {
+    ...buildCheck,
+    details: inspected.ok
+      ? {
+          reported_release_version: inspected.value.version,
+          launcher_sha256: inspected.value.sha256,
+          primary_fingerprint: inspected.value.primaryFingerprint,
+          source: inspected.value.source,
+        }
+      : { reason: inspected.error.reason, message: inspected.error.message },
+  };
 };
 
 const optionalHopperDemoCheck = async (
@@ -297,8 +314,8 @@ const providerDoctorChecks = (
 
 const unsupportedHopperRemediation = (configured: boolean): string =>
   configured
-    ? "Unset or update HOPPER_LAUNCHER_PATH, install the Hopper build supported by this REA release, or update REA."
-    : "Install the Hopper build supported by this REA release, or update REA for a newer Hopper build.";
+    ? "Check HOPPER_LAUNCHER_PATH and its publisher verification evidence. Run rea setup to install and retain an authenticated official Hopper package for offline use."
+    : "Run rea setup to install and retain an authenticated official Hopper package for offline use.";
 
 const parseMajor = (version: string | undefined): number =>
   Number.parseInt(version?.split(".")[0] ?? "0", 10);

src/application/DoctorProviderIds.ts

@@ -1,2 +1,8 @@
 /** Deep-analysis providers whose readiness can be selected explicitly by doctor. */
-export const DOCTOR_PROVIDER_IDS = ["hopper", "ghidra", "ida", "jadx"] as const;
+export const DOCTOR_PROVIDER_IDS = [
+  "hopper",
+  "ghidra",
+  "ida",
+  "jadx",
+  "kuna",
+] as const;

src/application/HermesConfigurationYaml.test.ts

@@ -8,6 +8,34 @@ import {
 } from "./HermesConfigurationYaml.js";
 
 describe("Hermes configuration YAML", () => {
+  it("retains a quoted sequence element when replacing its registration", () => {
+    const source =
+      "mcp_servers:\n  rea:\n    command: old\n    args:\n      - 'mcp'\n";
+    expect(
+      editHermesConfigurationYaml(source, ["mcp_servers", "rea"], {
+        command: "rea",
+        args: ["mcp"],
+      }),
+    ).toBe(source.replace("command: old", "command: rea"));
+  });
+
+  it("retains explicit string tags on unchanged keys and values", () => {
+    const source =
+      "%YAML 1.2\n---\nmcp_servers:\n  rea:\n    command: old\n    env:\n      !!str ON: !!str off\n";
+    const written = editHermesConfigurationYaml(
+      source,
+      ["mcp_servers", "rea"],
+      {
+        command: "rea",
+        env: { ON: "off" },
+      },
+    );
+    expect(written).toBe(source.replace("command: old", 'command: "rea"'));
+    expect(
+      parse(written.slice("%YAML 1.2\n---\n".length), { version: "1.1" }),
+    ).toMatchObject({ mcp_servers: { rea: { env: { ON: "off" } } } });
+  });
+
   it("writes an edited number from its new value", () => {
     expect(
       editHermesConfigurationYaml("mode: 0600\nmask: 0x1F\n", ["mode"], 420),

src/application/HermesConfigurationYaml.ts

@@ -1,9 +1,13 @@
 import {
+  isMap,
+  isNode,
   isScalar,
+  isSeq,
   parseDocument,
   Document,
   Schema,
   type ParsedNode,
+  type Node,
   type ScalarTag,
   type Tags,
 } from "yaml";
@@ -322,6 +326,36 @@ export const parseHermesConfigurationYaml = (text: string): unknown => {
       });
 };
 
+/** Keep unchanged scalar nodes, including their quoting and explicit tags. */
+const retainHermesScalarSource = (previous: unknown, next: Node): Node => {
+  if (isScalar(previous) && isScalar(next) && previous.value === next.value)
+    return previous;
+  if (isMap(previous) && isMap(next)) {
+    const entries = new Map(
+      previous.items.flatMap((pair) =>
+        isScalar(pair.key) && typeof pair.key.value === "string"
+          ? [[pair.key.value, pair] as const]
+          : [],
+      ),
+    );
+    for (const pair of next.items) {
+      const key = isScalar(pair.key) ? pair.key.value : undefined;
+      const prior = typeof key === "string" ? entries.get(key) : undefined;
+      if (prior === undefined) continue;
+      pair.key = prior.key;
+      if (isNode(pair.value))
+        pair.value = retainHermesScalarSource(prior.value, pair.value);
+    }
+  } else if (isSeq(previous) && isSeq(next)) {
+    next.items = next.items.map((item, index) =>
+      isNode(item)
+        ? retainHermesScalarSource(previous.items[index], item)
+        : item,
+    );
+  }
+  return next;
+};
+
 /** Set or delete one value in a Hermes `config.yaml`, keeping the rest as written. */
 export const editHermesConfigurationYaml = (
   text: string,
@@ -348,8 +382,26 @@ export const editHermesConfigurationYaml = (
       if (!document.hasIn(parent))
         document.setIn(parent, document.createNode({}));
     }
-    document.setIn(path, value);
+    const previous = document.getIn(path, true);
+    // Let YAML's scalar update preserve numeric formats for a direct edit.
+    // Replacing containers needs explicit retention of unchanged child nodes.
+    document.setIn(
+      path,
+      isMap(previous) || isSeq(previous)
+        ? retainHermesScalarSource(previous, document.createNode(value))
+        : value,
+    );
   }
   // Editing REA must not reflow unrelated user prompts or other YAML scalars.
-  return document.toString({ lineWidth: 0 });
+  return document.toString({
+    lineWidth: 0,
+    // Hermes tries the YAML 1.1 C loader before its directive-aware pure
+    // fallback. Quote new strings to avoid introducing YAML 1.1 coercions.
+    ...(document.directives?.yaml.version === "1.2"
+      ? {
+          defaultStringType: "QUOTE_DOUBLE" as const,
+          defaultKeyType: "QUOTE_DOUBLE" as const,
+        }
+      : {}),
+  });
 };

src/application/HopperDownload.ts

@@ -4,9 +4,23 @@ export const HOPPER_DOWNLOAD_BYTES = 100_000_000;
 /** Bound observed response bytes before retaining them, including chunked bodies. */
 export const downloadHopperPackage = async (
   url: string,
-  options: { readonly signal?: AbortSignal } = {},
+  options: {
+    readonly signal?: AbortSignal;
+    readonly maxBytes?: number;
+    readonly timeoutMs?: number;
+  } = {},
 ): Promise<{ readonly ok: boolean; readonly bytes: Uint8Array }> => {
-  const deadline = AbortSignal.timeout(120_000);
+  const timeoutMs = options.timeoutMs ?? 120_000;
+  if (!Number.isSafeInteger(timeoutMs) || timeoutMs < 1 || timeoutMs > 120_000)
+    throw new RangeError("Invalid Hopper response time budget");
+  const deadline = AbortSignal.timeout(timeoutMs);
+  const maxBytes = options.maxBytes ?? HOPPER_DOWNLOAD_BYTES;
+  if (
+    !Number.isSafeInteger(maxBytes) ||
+    maxBytes < 1 ||
+    maxBytes > HOPPER_DOWNLOAD_BYTES
+  )
+    throw new RangeError("Invalid Hopper response byte budget");
   const response = await fetch(url, {
     headers: { accept: "application/json", "user-agent": "rea-installer" },
     signal:
@@ -17,7 +31,7 @@ export const downloadHopperPackage = async (
   const refused = { ok: false, bytes: new Uint8Array() };
   if (
     !response.ok ||
-    Number(response.headers.get("content-length")) > HOPPER_DOWNLOAD_BYTES
+    Number(response.headers.get("content-length")) > maxBytes
   ) {
     await response.body?.cancel();
     return refused;
@@ -30,7 +44,7 @@ export const downloadHopperPackage = async (
     while (true) {
       const chunk = await reader.read();
       if (chunk.done) return { ok: true, bytes: Buffer.concat(chunks, bytes) };
-      if (chunk.value.byteLength > HOPPER_DOWNLOAD_BYTES - bytes) {
+      if (chunk.value.byteLength > maxBytes - bytes) {
         await reader.cancel();
         return refused;
       }

src/application/HopperReleaseFeed.test.ts

@@ -0,0 +1,77 @@
+import { describe, expect, it } from "vitest";
+
+import { parseLinuxHopperReleases } from "./HopperReleaseFeed.js";
+
+const release = {
+  version: "99.2.1",
+  filename:
+    "https://www.hopperapp.com:443/downloader/public/Hopper-99.2.1-Linux-demo.deb",
+  file_length: "123",
+  file_hash: "a".repeat(40),
+  main_exe_hash: "b".repeat(64),
+  signature: "AQ==",
+  public: "1",
+  beta: "0",
+  published_at: "preserved vendor field",
+};
+
+describe("Linux Hopper release feed boundary", () => {
+  it("preserves future release identity and partial platform facts", () => {
+    const parsed = parseLinuxHopperReleases({
+      "Ubuntu / Mint": release,
+      "OS X": { signature: "Sparkle DSA" },
+    });
+    expect(parsed).toMatchObject({
+      ok: true,
+      value: {
+        deb: { ok: true, value: release },
+        rpm: { ok: false },
+        arch: { ok: false },
+      },
+    });
+    if (!parsed.ok || !parsed.value.deb.ok) throw new Error("Missing release");
+    expect(parsed.value.deb.value.filename).toBe(release.filename);
+  });
+
+  it.each([
+    { signature: undefined },
+    { signature: "AQ==garbage" },
+    { main_exe_hash: "b".repeat(40) },
+    { file_length: "100000001" },
+    { file_length: "NaN" },
+    { public: "0" },
+    { beta: "1" },
+    {
+      filename:
+        "https://other.example/downloader/public/Hopper-99.2.1-Linux-demo.deb",
+    },
+    { filename: release.filename + "?token=caller" },
+    {
+      filename: release.filename.replace("https://", "https://user:password@"),
+    },
+    { filename: release.filename.replace(".deb", ".rpm") },
+    { filename: release.filename.replace("demo.deb", "full.deb") },
+  ])(
+    "rejects metadata outside the selected public package contract: %j",
+    (changes) => {
+      expect(
+        parseLinuxHopperReleases({
+          "Ubuntu / Mint": { ...release, ...changes },
+        }),
+      ).toMatchObject({
+        ok: true,
+        value: { deb: { ok: false, error: { reason: "release_metadata" } } },
+      });
+    },
+  );
+
+  it.each([null, [], "html error page"])(
+    "keeps malformed feed roots unknown",
+    (raw) => {
+      expect(parseLinuxHopperReleases(raw)).toMatchObject({
+        ok: false,
+        error: { reason: "release_metadata" },
+      });
+    },
+  );
+});

src/application/HopperReleaseFeed.ts

@@ -0,0 +1,125 @@
+import { z } from "zod";
+
+import { err, ok, type Result } from "../domain/result.js";
+import { HOPPER_DOWNLOAD_BYTES } from "./HopperDownload.js";
+
+/** Vendor endpoints, shared by installation and existing-build inspection. */
+export const HOPPER_RELEASES_URL =
+  "https://www.hopperapp.com/include/files-api.php?request=releases&public=true";
+export const HOPPER_PUBLIC_KEY_URL =
+  "https://www.hopperapp.com/hopper-public-key.asc";
+export const HOPPER_DOWNLOAD_PREFIX =
+  "https://www.hopperapp.com:443/downloader/public/";
+
+/** Stable publisher identity; signing subkeys are admitted through this primary. */
+export const HOPPER_LINUX_PRIMARY_FINGERPRINT =
+  "7C2D73BAFAF5B9A710A69DA1A26DF815099B1BCD";
+
+export type LinuxPackageFamily = "deb" | "rpm" | "arch";
+const platforms = {
+  deb: "Ubuntu / Mint",
+  rpm: "Fedora",
+  arch: "Arch",
+} as const;
+const suffixes = { deb: ".deb", rpm: ".rpm", arch: ".pkg.tar.xz" } as const;
+
+export class HopperReleaseError extends Error {
+  readonly _tag = "HopperReleaseError";
+  constructor(
+    readonly reason:
+      | "release_metadata"
+      | "download"
+      | "integrity"
+      | "publisher_verification_failed"
+      | "verification_unavailable"
+      | "unsupported_hopper_build"
+      | "cancelled"
+      | "cleanup_failed"
+      | "cache_missing",
+    message: string,
+    options?: ErrorOptions,
+  ) {
+    super(message, options);
+    this.name = this._tag;
+  }
+}
+
+const sha256 = z.string().regex(/^[a-f0-9]{64}$/u);
+const releaseSchema = z
+  .object({
+    version: z.string().min(1),
+    filename: z.string().url(),
+    file_length: z
+      .string()
+      .regex(/^[1-9]\d*$/u)
+      .refine((text) => {
+        const count = Number(text);
+        return Number.isSafeInteger(count) && count <= HOPPER_DOWNLOAD_BYTES;
+      }),
+    file_hash: z.string().regex(/^[a-f0-9]{40}$/u),
+    main_exe_hash: sha256,
+    signature: z
+      .string()
+      .min(1)
+      .refine(
+        (text) => Buffer.from(text, "base64").toString("base64") === text,
+      ),
+    public: z.literal("1"),
+    beta: z.literal("0"),
+  })
+  .passthrough();
+
+/** Parsed once at the feed boundary; preserve the vendor's complete release row. */
+export type LinuxHopperRelease = z.infer<typeof releaseSchema>;
+export type LinuxHopperReleaseFeed = Readonly<
+  Record<LinuxPackageFamily, Result<LinuxHopperRelease, HopperReleaseError>>
+>;
+
+export const parseLinuxHopperRelease = (
+  raw: unknown,
+  family: LinuxPackageFamily,
+): Result<LinuxHopperRelease, HopperReleaseError> => {
+  const parsed = releaseSchema.safeParse(raw);
+  if (!parsed.success)
+    return err(
+      new HopperReleaseError(
+        "release_metadata",
+        "The Hopper release omitted valid package identity or signature metadata.",
+      ),
+    );
+  const url = new URL(parsed.data.filename);
+  if (
+    url.origin !== "https://www.hopperapp.com" ||
+    url.username !== "" ||
+    url.password !== "" ||
+    url.search !== "" ||
+    url.hash !== "" ||
+    !url.pathname.startsWith("/downloader/public/Hopper-") ||
+    !url.pathname.endsWith(`-Linux-demo${suffixes[family]}`)
+  )
+    return err(
+      new HopperReleaseError(
+        "release_metadata",
+        "The Hopper release is outside the official public Linux demo channel.",
+      ),
+    );
+  return ok(parsed.data);
+};
+
+export const parseLinuxHopperReleases = (
+  raw: unknown,
+): Result<LinuxHopperReleaseFeed, HopperReleaseError> => {
+  const feed = z.record(z.string(), z.unknown()).safeParse(raw);
+  if (!feed.success)
+    return err(
+      new HopperReleaseError(
+        "release_metadata",
+        "The Hopper releases response is not an object.",
+      ),
+    );
+  return ok({
+    deb: parseLinuxHopperRelease(feed.data[platforms.deb], "deb"),
+    rpm: parseLinuxHopperRelease(feed.data[platforms.rpm], "rpm"),
+    arch: parseLinuxHopperRelease(feed.data[platforms.arch], "arch"),
+  });
+};

src/application/InvestigationProviders.ts

@@ -18,6 +18,7 @@ export const workflowAnalysisProfile = (
 ): AnalysisProfileCommitment =>
   createAnalysisProfile(REA_WORKFLOW_PROVIDER, {
     upstream_analysis_profile: upstream,
+    source_limitations: "provider-operation-caveats-v1",
     ...(operation === "binary_overview"
       ? { workflow_revision: 4 }
       : operation === "batch_decompile" ||