mirror of
https://github.com/morluto/rea.git
synced 2026-11-16 19:53:50 +00:00
ff0176256a98f0eac8c8fc0aff511544073c8a98
9
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
|
|
ff0176256a |
feat(snapshots)!: persist the target identity as the snapshot target (#11551)
* feat(snapshots)!: persist the target identity as the snapshot target
The snapshot target is now `{ kind, identity }`, where `identity` is the
FileTargetIdentity that `binaryTargetIdentity` produces and `kind` is the
binary kind query IDs commit to. The per-format target projection and
`snapshotTargetIdentity` are deleted; the schema checks the format's
selection, architecture and kind instead of keeping one union arm per format.
Entry execution subjects use the Evidence subject shape (file or explicit
unknown), so a snapshot entry commits to the same identity, including the
selection, as its Evidence record. Workflow entries take the subject of the
workflow's own Evidence. Replay no longer re-derives the selection from the
open target.
`FileTargetSelection` gains `universal-slice` (`offset`, `size`, and the FAT
`cpu_type`/`cpu_subtype` when target admission supplied the FAT table, else
null). `binaryTargetIdentity` produces it for `mach-o-universal` targets, so
Evidence and snapshots for different slices of one container no longer share
an identity.
Refs #11417
BREAKING CHANGE: analysis snapshot files change shape: `target` is
`{ kind, identity: { kind: "file", sha256, format, architecture, selection } }`
and entry `execution.subject` is the Evidence subject
(`{ identity, name, local_path }`, never null). Snapshot binding mismatch
details report `target.snapshot`/`target.selected` in the same shape. Evidence
IDs of universal Mach-O targets change because their subject identity now
carries the selected slice. Recreate saved snapshots and Evidence bundles of
universal Mach-O targets.
* docs(evidence): name the universal Mach-O slice selection in the subject contract
* fix(snapshots): read the snapshot target identity in the Ghidra verification scripts
The real Ghidra snapshot checks (lifecycle, RISC-V, MIPS, PowerPC) read
snapshot.target.sha256 and .architecture, which moved under
snapshot.target.identity, so they would fail after saving a valid snapshot.
* feat(evidence)!: name live browser page Evidence by its page URL (#11559)
Browser page observations (inspect_web_page, analyze_web_bundle,
observe_web_session, discover_webmcp_tools, capture_web_screenshot,
observe_web_execution, inspect_web_event_listeners) named an unknown
(`not-observed`) subject although the provider reports the page URL.
`TargetIdentity` gains a `url` kind: the reported URL with userinfo removed
and its origin (`null` for an opaque origin). The schema refuses a URL that
carries userinfo or an origin the URL does not have.
`browserUrlTargetIdentity` in the browser URL owner builds it from the
provider's reported URL; a withheld or unparseable URL stays unknown
(`not-observed`). Target lists and comparisons still name no single
document. Analysis views keep their parent's URL subject, as they already
keep a file subject.
BREAKING CHANGE: Evidence IDs of live browser page observations and their
analysis views change, and their subject is
`{ "kind": "url", "url", "origin" }` with `name` and `local_path` null.
Recreate stored Evidence for these tools. This lands inside Evidence format
revision 2 under the release hold (#11417); no `evidence_format` bump.
|
||
|
|
d87be930c4 |
refactor(providers)!: key Ghidra and Hopper analysis profiles on the producer identity instead of hand-bumped markers (#11320)
* refactor(ghidra)!: key analysis profiles on the producer identity instead of hand-bumped markers Ghidra snapshot validity depended on 53 hand-maintained "-vN" semantic markers in the analysis profile; a behaviour fix that forgot to bump its marker left old snapshots admissible (#3664). #10562 added a digest of the bridge sources, but TypeScript projection changes still needed a bump and the REA release was not committed. - src/filesystem/ProducerIdentity.ts owns the producer identity: the REA release plus a framed SHA-256 of a producer's packaged scripts, read once per process. - Ghidra profiles commit `producer` in place of `bridge_sources_sha256` and the constant revision markers (shared semantics block, load-image, entry seed, PDB, jump-table, DOL import, MIPS ISA-mode markers) are removed. - Runtime-selected values keep their meaning without a revision suffix (language_selection, annotation_policy, relocation/no-return/launch policy, MIPS/PowerPC/PSP lanes, MIPS ABI basis). Bridge wire policy ids shared with the Java scripts (ilp32-prototype-v1, o32-soft-float-prototype-v1) are unchanged. BREAKING CHANGE: Ghidra Evidence `analysis_profile.parameters` replace the semantic marker keys and `bridge_sources_sha256` with `producer` ({rea_version, sources_sha256}), and selection values lose their "-vN" suffix (for example `language_selection: "provider-pinned"`). Ghidra Evidence IDs and profile digests change, and analysis snapshots saved by earlier builds are refused at open. Migration: open the target without the old snapshot and save a fresh one; consumers reading the removed marker keys should read `producer` instead. * fix(hopper)!: commit the REA release and bridge script in Hopper analysis profiles (#11324) Hopper analysis profiles committed the launcher digest and one hand-bumped marker (function_dossier_coverage "...-v1"), but neither the REA release nor bridge/hopper_bridge.py. A snapshot saved before a bridge or REA change replayed results the current bridge would not produce. HopperBridgeSources owns the packaged bridge script URL (used to launch it) and hopperProducerIdentity() on the shared ProducerIdentity owner; the profile commits `producer` in place of the marker. BREAKING CHANGE: Hopper Evidence `analysis_profile.parameters` replace `function_dossier_coverage` with `producer` ({rea_version, sources_sha256}). Hopper profile digests and Evidence IDs change, and analysis snapshots saved by earlier builds are refused at open. Migration: open the target without the old snapshot and save a fresh one. * fix(ghidra): build the legacy reference snapshot from another producer and use the shared temp directory * refactor(providers): key Kuna exports and the Native AOT profile on the producer identity (#11353) * fix(ghidra): keep the bridge source table import-free for the Node-loaded verifier scripts/verify-ghidra-bridge-sources.mjs loads GhidraBridgeSources.ts with Node type stripping, which does not resolve .js specifiers to .ts sources. The producer identity import made every verifier run fail with ERR_MODULE_NOT_FOUND; the identity now lives in GhidraProducerIdentity.ts. * fix(artifacts): import the moved file source and abort check from their resources owner after merging main |
||
|
|
15107aa81b | test(ghidra): check structured snapshot profile mismatch (#8536) | ||
|
|
7d36d172ab | fix(evidence): consolidate retained view and verifier ownership | ||
|
|
1a48a82020 |
feat(analysis)!: default function MCP output to retained summaries (#5660)
* feat(ghidra): bound one procedure's decompile without closing the session Fixes #5340 * fix(ghidra): report decompile limit as a request timeout Keeps BinarySession from marking every Ghidra operation unavailable after one procedure's decompile reaches the limit, and names the MCP tool-call deadline in the remediation. * fix(ghidra): budget the decompile limit and report cancellation-closed sessions Adds REA_GHIDRA_DECOMPILE_TIMEOUT_SECONDS to the derived Codex/Kimi tool deadline and forwards it through setup. Marks a remote failure that settles after caller cancellation so the timeout projection no longer promises session reuse. * feat(ghidra)!: default analyze_function to a retained summary analyze_function was the one analysis tool never wired into REA's detail: summary|complete convention, so it always returned the complete native dossier inline (0.2-10 MB, dominated by value-flow and references). Agents on tool-result-capped harnesses never reached the pseudocode. Add `detail` to analyze_function, defaulting to summary: return a compact dossier (procedure identity, pseudocode size, facet counts, value-flow availability) and retain the complete dossier Evidence in the session. Page pseudocode, references, assembly and value-flow through normalized_result.parent_evidence_id with inspect_analysis_view, which already projects every native facet. detail: complete returns the full dossier inline as before. Delivery detail never reaches the provider or Evidence identity. BREAKING CHANGE: analyze_function now defaults to the compact summary view instead of the complete dossier inline. Pass detail: "complete" for the previous behavior. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * feat(cli): expose analyze_function detail on the function command Wire `rea function` into the shared CLI detail-delivery convention (--detail, --evidence_output) so the CLI and MCP adapters offer the same summary | complete choice for analyze_function. Summary delivery writes the complete dossier Evidence to a bundle and returns the compact summary view; it fails before analysis when --evidence_output is missing, so no provider work is wasted. The CLI default stays complete because a one-shot command has no in-process Evidence retention. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * refactor(ghidra): harden analyze_function summary after independent review FlatlineRoundtable cross-model review (6 lanes) converged on the summary projection's availability semantics being underspecified. Document that facet_availability is authoritative over raw pseudocode length, that code_units are UTF-16 (the unit inspect_analysis_view pages by), and that value_flow.truncated is meaningful only when value-flow is available. Parenthesize the nullish-coalescing/comparison for clarity. No behavior change; declined claims (value-flow "truncated collapse", unresolved_calls dead code) were schema-impossible or already correct. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * test(ghidra): cover analyze_function detail wiring on both adapters Close the roundtable's top convergent finding: the summary path had only domain-projection coverage, not the server/CLI wiring where the breaking behavior lives. MCP server (registerEnhancedTools): the default delivers a retained summary whose complete parent is recorded for paging; delivery detail is stripped before the provider request; detail: complete returns the full dossier inline; a registration without retention rejects a summary request before any provider work. CLI (rea function): the default returns the complete dossier; --detail summary without --evidence_output fails before analysis; with the bundle sink it returns the compact summary and writes the complete Evidence bundle. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * fix(ghidra): keep analyze_function delivery detail out of provider inputs Address the five automated review findings on #5660. P1: `detail` lived in the shared enhancedInputSchemas.analyze_function, so inspect_native_api composed `{procedure, detail: "summary"}` and the strict Ghidra analyze_function schema rejected it. Delivery detail is now owned by the analyze_function MCP tool contract; the handler separates it before any provider call, and executeFunctionAnalysisEvidence parses the provider input only. The MCP regression test validates provider input exactly as the production Ghidra client does and exercises inspect_native_api. P2: the analyze_function summary Evidence now carries the parent's relationship_coverage like native facet views, so partial or unknown caller/callee discovery is not reported as complete. P2: the summary reports the dossier's declared facet_availability, nulls the counts of facets the provider declared unavailable, and lists each reason in unknowns, so Kuna's empty references, comments and basic blocks are not mistaken for observed absence. P2: inspect-analysis-view evidence-file sources accept the Evidence bundle that --evidence_output writes, selecting a record with source.evidence_id when the bundle holds several; the CLI test reads the saved bundle back through that route. Tests that assert the complete dossier over public analyze_function now pass detail complete, and the oversized-dossier test asserts the summary view it previously expected to be unsupported. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * fix(ghidra): keep omitted function facets unknown in the summary The analyze_function summary coerced an omitted unresolved_calls record to a count of 0 while the native facet view reports the same dossier as "not recorded", and it carried no counts at all for the pageable swift_string_literals facets. Count optional records as null when the dossier omits them, name them in unknowns, and add nullable counts for both Swift literal facets. The summary Evidence envelope also dropped the function address: createAnalysisViewEvidence now derives one function-view predicate for both the address location and relationship coverage. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * fix(scripts): request complete analyze_function dossiers in real-provider checks The real-provider check scripts call analyze_function over MCP and then consume the complete dossier: requireFunctionDossier, functionDossierSchema, .procedure/.pseudocode reads, operation assertions, compare_functions inputs and the oversized-dossier rejection path. The summary default broke them. Every such call site now passes detail: "complete". Provider-direct callers are unchanged because detail is an MCP contract field; the admission-error sites stay unchanged because provider failures return before the summary branch. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * fix(ghidra): keep unavailable value-flow counts unknown in the summary summarizeDossier coerced the five value_flow_* counts to zero when native_value_flow was null (IDA, Kuna) or declared unavailable (Ghidra), so the summary reported empty collections while value_flow.available said otherwise. The counts are now null unless value flow is available, with the reason already named in unknowns. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * fix(evidence): budget authentication headroom by file shape and serialized size The evidence-file headroom reserved two parsed copies, calibrated for one record, and charged strings at two bytes per character. Bundle authentication additionally holds the schema-parsed bundle, the re-parsed records and two JSON.stringify results, which expand control characters and lone surrogates sixfold: a bundle whose raw_result is a million NUL characters estimated at 2 MB parsed but needed 12 MB of serialized text. The budget now has a parsed term and a serialized term per file shape, owned by EvidenceAuthenticationHeadroom and shared by the analysis-view reader and the bundle importer, which had no guard. Two transient peaks the budget did not name are removed at their producers: the canonical digest hashes a large leaf directly instead of concatenating it into its buffer, and the JSON validator walks containers by cursor so a wide array no longer queues one entry per member. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * fix(mcp): route analyze_function summary replies to dossier consumers in docs and prompts The MCP guide, the version-comparison and crash-triage prompt contracts, the authored skill and the Kuna, IDA and function-matching docs still told agents to pass analyze_function replies straight to compare_functions or build_call_path. Under the summary default those consumers reject the reply. They now name the retained-evidence reference for compare_functions, the observations fields of match_native_functions, and detail: "complete" for build_call_path, which accepts only inline dossiers. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * test(evidence): pin validator member order and keep interrupted validation yielding Independent review of the cursor validator found the interrupted-validation fixture relied on the old last-first walk: its over-deep head was reached before the first cooperative yield. The padding now precedes the deep element so the walk yields first, which is the invariant the test guards. A new case pins first-defect-in-member-order reporting, skipped non-enumerable keys no longer count toward the yield cadence, and the import-headroom test loads the reader through a fresh module graph so it holds under the services project's shared workers. Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01AoJZDPcrgmMsCqLG4LfiFb * fix(ghidra): align deadline bounds and cancellation outcomes * refactor(ghidra): shorten deadline recovery guidance * fix(mcp): clarify the scope of inferred decompile deadlines * test(ghidra): verify deadline workflows and document measured costs Separate failure and registration cases by their distinct contracts. Record actual Linux Ghidra, CLI and MCP evidence with measured context, elapsed time, CPU and memory, and explicit remaining coverage limits. * docs(ghidra): pin integration and distinguish baseline format failures * fix(ghidra): own failure lifecycle wording in the provider * test(ghidra): budget timeout fixture generation before allocation * fix(analysis): preserve summary provenance and bound bundle guidance * fix(ghidra): align capability and traversal deadline guidance * style(config): align deadline guidance formatting * docs(ghidra): record capability and traversal verification * docs(analysis): use absolute paths for Evidence files * docs(analysis): record completed function summary measurements * docs(analysis): update main integration validation --------- Co-authored-by: Syed Rishad <rishadzzzz@gmail.com> Co-authored-by: Claude Opus 4.8 <noreply@anthropic.com> Co-authored-by: gensokyo-dev <342580529+gensokyo-dev@users.noreply.github.com> |
||
|
|
f3c4a332f7 | fix(targets): cancel resolution before provider admission | ||
|
|
7e96f25f8b |
fix(mcp): deliver typed errors without invalid structured output (#1277)
* fix(mcp): deliver tool errors within advertised schema contracts Retain complete typed JSON error content while omitting non-conforming structured errors at the public transport boundary. Recover oversized diagnostics into Evidence before removing their private structured carrier. Update schema-validated client readers and cover actual stdio delivery. Closes #1261 Assisted-by: Kogasa (gensokyo-skills:kogasa-surprise-testing) * test(mcp): read typed errors in canonical Evidence regressions * test(mcp): validate public error diagnostics in acceptance workflows |
||
|
|
63bf91405c |
refactor!: enforce canonical evidence and provider boundaries
Deliver complete Evidence directly through MCP with required nullable analysis profiles and immutable per-server delivery settings. Bind selected environments at construction and require typed cleanup outcomes with retained retry ownership. Remove test-only routing, duplicate session and snapshot metadata, producer-field repairs, and ambiguous provider arguments. Separate artifact byte identity from occurrence classification and retain typed source-map and partial native facts across CLI, MCP, and verifier consumers. BREAKING CHANGE: obsolete Evidence envelopes, session aliases, incomplete snapshots, and missing normalized producer facts are no longer accepted. |
||
|
|
84a17d5519 |
fix(ghidra): preserve function references and snapshot boundaries
Collect references from the complete owned function body and exact entry, including instruction-interior destinations and embedded-data sources. Bind the corrected semantics to the analysis profile so legacy snapshots cannot replay incomplete results. Preserve snapshot mismatch reasons and recovery advice through CLI and MCP without disturbing live sessions. Return exact entry candidates for ambiguous function names. Expand the real-provider regression workflow and replace redundant parser assertions with malformed-output coverage through the pinned MCP client. Verified with Ghidra 12.1.4 on macOS arm64: full host-native lane, real CLI/MCP boundary probes, 30 focused tests, and typecheck/lint. |