Commit Graph
14 Commits
Author SHA1 Message Date
morluto 332dea76a6 fix(native): keep unreported value-flow and archive coverage unknown, and exempt generation caps by reason marker (#11288)
* fix(native): keep unreported value-flow truncation and archive decode coverage unknown

* test(verify): count multiline optional-fact defaults and pin the guard's exclusions

* fix(domain): keep unknown counts and costs unknown in comparison and coverage summaries (#11325)

* fix(domain): keep unknown counts and costs unknown in comparison and coverage summaries

* refactor(verify): exempt legitimate optional-fact defaults with a reasoned marker (#11355)

* refactor(verify): exempt legitimate optional-fact defaults with a reasoned marker

The optional-fact-default guard counted request defaults and absence that
is modelled elsewhere as hidden unknowns. A comment
`// optional-fact-default: <reason>` on the default's line or the line
before now exempts it; the guard reads unstripped source for the marker and
ignores matches inside comment lines.

Marked: the two evidence paging offsets and four Swift literal reads whose
absence swiftLiteralCoverage reports. Removed instead of marked: jsdiff 9
types Change.count as required, so its two `?? 0` defaults were dead; the
evidence link traversal now carries retained records, so it no longer
re-looks them up behind `?? []`. Baseline 31 -> 22.

* fix(verify): blank comment spans instead of skipping comment-led lines in the optional-fact guard

Skipping a match whose line starts like a comment hid live code after a
leading block comment and counted defaults inside a block comment without
leading stars. The counter now blanks comment text (keeping offsets and line
breaks, skipping string literals) and matches the blanked code, while
exemption markers are still read from the original lines.

* refactor(domain): mark optional-fact defaults whose absence is not an observation (#11389)

* refactor(domain): mark optional-fact defaults whose absence is not an observation

Per-site review of the remaining optional-fact-default entries. Each marked
default is a caller option (trace ignore_fields, environment rules), a list
whose producer omits the empty case (unknown_size_frames, the schema's
min-1 representation_node_ids), caveat text whose completeness another field
carries (call diagnostics, unresolved calls, semantic resource limits), or
absence that already turns the result unknown (Wasm interface inventory,
WebMCP discovery, managed ExportedType rows, legacy integrity mismatch IDs).

Left counted: webCaptureDiff redirects (2; the CDP producer always records
them, so absence is a legacy record that should compare as unknown, not
as no redirects) and HopperCallerReconciliation (2; index-miss semantics not
yet verified). Baseline 22 -> 4.

* fix(browser): compare redirects as unknown when a capture predates redirect recording (#11402)

* fix(browser): compare redirects as unknown when a capture predates redirect recording

The CDP producer records every request's redirect chain, empty when there was none; a record without one predates redirect recording (#986). Comparing it against a current capture reported the network dimension as unchanged, treating the missing chain as "no redirects". Redirects are now compared only when both captures recorded them, and the network dimension is otherwise unknown with a reason; proven membership changes are still reported.

* refactor(hopper): interpret a caller-index miss once at the lookup (#11403)

The forward index holds observations from the examined procedures only, and relationship coverage stays unknown with the scan facts as its reason. A miss therefore adds no callers without showing there are none; the lookup now says so once instead of four optional reads defaulting at each site. The optional-fact-default baseline is now empty.

* docs(javascript): state why an omitted representation list adds no nodes

* fix(verify): count optional-fact defaults from parsed TypeScript, not comment-blanked text

The guard blanked comments with a hand-written scanner and matched a regular expression on the rest, so a default quoted in a string counted and a regular expression holding /* hid the code after it. It now walks the TypeScript syntax tree for property reads defaulted with ?? 0, ?? [] or ?? false, and reads exemption markers only from comment trivia.

* fix(verify): tokenize optional-fact defaults without a compiler and count chained defaults

* fix(verify): match whole optional-fact default expressions across calls and operators

* fix(verify): count optional-fact defaults in template substitutions and after postfix updates, and tie markers to their own expression

* refactor(verify): exempt generation and sample caps from the constant-maximum-slice guard by reason marker (#11454)

* refactor(verify): exempt generation and sample caps from the constant-maximum-slice guard by reason marker

* fix(verify): read reason markers only from real line comments

* fix(verify): read reason markers from the shared source tokenizer

* fix(verify): read a regular expression after a control head as a statement start

The shared tokenizer treated every closing parenthesis as the end of an operand,
so in `if (ready) /[/*]/.test(v)` it read `/[` as division and `/*` inside the
regex as a block comment, hiding the rest of the file from the marker-aware
guards. Parentheses are now tracked: the `)` closing an if/while/for/with head
allows a regular expression, while a call's `)` still means division.

* fix(verify): keep a keyword-named member call out of control-head detection

`policy.if(ready) / (…)` was read as an if-statement head, so the division was
taken for a regular expression and the denominator's optional-fact default was
not counted. A control keyword after `.` or `?.` is a member call.

* refactor(verify): exempt budget-arithmetic cursors and tallies by a not-an-allowance marker (#11493)

* refactor(verify): exempt budget-arithmetic cursors and tallies by a not-an-allowance marker

The budget-arithmetic guard cannot tell a scan cursor, an observed total, a
string being assembled or a tail buffer's retained length from an allowance
when each is compared with a limit. A field declaration marked
`// not-an-allowance: <reason>` (read through the shared reason-marker scan)
is no longer counted. The four remaining sites carry a marker, so the
baseline is empty and removed.

* refactor(verify): exempt detail levels, lifecycle states and domain vocabularies from the completeness-enum guard by reason marker (#11510)

A zod enum containing "complete" is counted as a local coverage vocabulary. Eleven
are not coverage: a caller-chosen detail level, how far a command or session
ran, and format states (GPU decode depth, WASM name section, SQLite column
details). They now carry `// detail-level:`, `// lifecycle-state:` or
`// domain-vocabulary:` with a reason, read through the shared reason-marker
scan (which gains an optional predicate for the matches a guard counts).
local-completeness-enum: 22 -> 11 on this branch.

* fix(verify): bind a not-an-allowance marker to its own class field

Exemptions were a file-wide set of names, so a marked tally in one class hid an
unmarked allowance with the same name in another, and a marked object property
could exempt a class field. Markers now exempt only a member declaration (at
class member depth) of the top-level class that contains it, and the named-
allowance count is taken per class so the same scoping applies to it.

* fix(artifacts): keep main's resources import in the Deno standalone reader after merging main

* fix(verify): start an expression after a template substitution and a non-member keyword, and accept any marker reason

The shared tokenizer read the `/` after `${` as division, so a regular
expression holding `/*` in a substitution hid the rest of the file from
every tokenizer-based guard, and it treated property names such as
`row.return` as keywords. Tokens now record whether template text opens a
substitution and whether a name is an expression keyword (not after `.` or
`?.`), and the regular-expression rule and the optional-fact-default operand
rule read those facts. The optional-fact-default guard reads markers through
reason-marker's isReasonMarker, so any non-empty reason exempts, as for the
other reason markers.

* fix(verify): blank comments with the shared tokenizer before matching architecture guards (#11573)

* fix(verify): blank comments with the shared tokenizer before matching architecture guards

The guards stripped comments with a regular expression that took a `/*`
inside a string, template or regular expression for a comment opener and
deleted code up to the next `*/`. src/flutter/FlutterBuildProvider.ts's
"lib/*/libapp.so" text hid four --pool-policy remediation strings from
cli-option-text-outside-cli. blankComments in source-tokens.mjs now owns
comment removal for the architecture guards and the reason-marker scan: it
blanks comment tokens in place (newlines kept) and keeps strings, templates
and regular expressions. The scan behind it retains only the last tokens it
reads, and whitespace runs are skipped at once.

The cli-option-text-outside-cli baseline records FlutterBuildProvider.ts at
its true count of 4; no other baseline changed.

Fixes #11477

* fix(verify): read a regular expression after a statement block and keep comment offsets for wide characters

The tokenizer now records whether each `{` opens a statement block or an
object literal (from the token before it) and marks the `}` that closes a
block, after which a statement, and so a regular expression, can start; an
object literal's `}` still ends an operand. blankComments replaces each run
of comment text with one space per UTF-16 unit, so a surrogate pair no longer
shifts later offsets.

* chore(verify): drop the Flutter CLI-text baseline entry now that main names no option there
2026-11-16 07:40:30 +08:00
morluto 5c53bcaed8 feat(mcp)!: retire the full schema profiles and declare schema facts as typed annotations (#10693)
* test(mcp): check advertised schemas against a portable dialect in every profile

* fix(mcp): re-emit advertised patterns in a portable regex grammar

* fix(mcp): present discover_tools input through the selected profile

The hot-cold catalog registered discover_tools directly with the SDK, so
the default compact input profile still advertised its canonical schema,
including page-size defaults clients apply on their own (#5322). Route
every listed input schema, transport discovery included, through one
server-owned presentation, and pass the selected presentation to the
deferred server, which serves recorded tool schemas but not discovery.

* fix(mcp): keep the Evidence union on described borrowed inputs

borrowEvidenceInput installed its advertised projection on one schema
instance. A call site's .describe() rebuilds the schema from its
definition, so match_native_functions advertised left and right as
unconstrained values in every input profile. Install the projection in
the borrowed parser's constructor so derived schemas keep it, for both
borrowed inputs and borrowed arrays.

* fix(mcp): validate advertised dialect by selected profile

* fix(mcp): preserve schema widening through references

* feat(mcp): project compact input schemas into the portable advertised dialect

One projection (advertisedDialectProjection.ts) owns the compact input
advertisement: one type per node, items on every array, string literals
as enums, oneOf widened to anyOf, allOf merged, tuples as element unions,
exclusive bounds made inclusive, and every keyword outside the dialect
dropped. Each rewrite is exact or widening; server validation is
unchanged. The root description and sentences for dropped cross-field
constraints move to the tool description, which Codex and Zed keep.

The compact baseline falls to nullable values only (29), which the
dialect cannot express without narrowing an input. The gateway shape
repairs (typeArrayItemUnions, hasGatewayIncompatibleShape), the second
dependentRequired-to-prose copy and the strict-gateway test denylist are
removed; the dialect checker gates the inline presentation instead.

* feat(mcp): omit output schemas from tools/list by default

No surveyed model API reads an output schema, MCP client validators
repeat REA's own result validation, and complete output schemas were
about 92% of the default tools/list. REA_MCP_OUTPUT_SCHEMA_PROFILE now
resolves to off through one owner (resolveMcpOutputSchemaProfile);
compact and full remain selectable. structuredContent is still
delivered, and the deferred server serves both off and full from the
recorded catalog with the matching recorded identity.

Tests and artifact verifiers that validate results against complete
advertised output contracts select full explicitly.

* docs: describe omitted output schemas as the default in README translations

* fix(mcp): select schemas in verification workflows

* fix(mcp): refuse zero-padded repeat bounds and record host-specific path patterns

* fix(mcp): keep projected schemas permissive

* fix(mcp): keep references valid when presentation widens their targets

* fix(mcp): keep a single union branch's closure scoped to its properties

* fix(mcp): select complete output schemas in the Electron verifier

* fix(mcp): select complete output schemas in standalone verifier probes

* fix(mcp): keep an allOf member's tuple tail with its own positions

* fix(mcp): relax references into negations and conditionals presentation may remove

* refactor(contracts): declare root discriminators as typed schema annotations

* test(mcp): select the borrowed Evidence contract from its family

* fix(mcp): keep astral literals and dynamic references widening in every mode

* fix(mcp): omit astral literals and relax untracked references when presentation widens

* fix(mcp): select complete output schemas in every verifier transport

* fix(mcp): read reference pointers by schema position, not segment spelling

* fix(mcp): protect references into collapsing boolean unions and keep referencing branches out of the collapse

* fix(mcp): keep reduced schemas open when their root admits names by pattern

* test(mcp): keep the reduced-schema pattern case outside the size-limited block

* fix(mcp): omit patterns whose classes name astral characters

* fix(mcp): decide union collapse and pointer protection from canonical branches

* docs(mcp): document non-u astral matching and pin BMP agreement across V8 modes

* fix(mcp): keep prototype-named properties as own properties in the dialect projection

* test(mcp): select full output schemas in packet and signal schema checks

* test(mcp): expect the advertised base64 schema to omit the unbounded pad-bit probe

* fix(mcp): protect references into unevaluated keywords lost evaluation removes

* fix(mcp): validate the doctor probe reply against the canonical output contract (#10655)

* fix(mcp): validate the doctor probe reply against the canonical output contract

* test(mcp): replay doctor probe validation under every output profile

* test(mcp): accept typed-branch nullable unions in the advertised dialect (#10649)

* test(mcp): select full output schemas in HTTP session workflows

* fix(mcp): emit stack-safe classes with non-ASCII members and protect draft-07 reference siblings

* fix(mcp): protect references into literal siblings of relaxed draft-07 references

* feat(mcp): omit output schemas from tools/list by default (#10646)

* feat(mcp): omit output schemas from tools/list by default

No surveyed model API reads an output schema, MCP client validators
repeat REA's own result validation, and complete output schemas were
about 92% of the default tools/list. REA_MCP_OUTPUT_SCHEMA_PROFILE now
resolves to off through one owner (resolveMcpOutputSchemaProfile);
compact and full remain selectable. structuredContent is still
delivered, and the deferred server serves both off and full from the
recorded catalog with the matching recorded identity.

Tests and artifact verifiers that validate results against complete
advertised output contracts select full explicitly.

* fix(mcp): select schemas in verification workflows

* fix(mcp): select complete output schemas in the Electron verifier

* fix(mcp): select complete output schemas in standalone verifier probes

* fix(mcp): select complete output schemas in every verifier transport

* test(mcp): select full output schemas in packet and signal schema checks

* fix(mcp): validate the doctor probe reply against the canonical output contract (#10655)

* fix(mcp): validate the doctor probe reply against the canonical output contract

* test(mcp): replay doctor probe validation under every output profile

* test(mcp): accept typed-branch nullable unions in the advertised dialect (#10649)

* test(mcp): select full output schemas in HTTP session workflows

* fix(mcp): keep the Evidence union on described borrowed inputs (#10597)

* fix(mcp): keep the Evidence union on described borrowed inputs

borrowEvidenceInput installed its advertised projection on one schema
instance. A call site's .describe() rebuilds the schema from its
definition, so match_native_functions advertised left and right as
unconstrained values in every input profile. Install the projection in
the borrowed parser's constructor so derived schemas keep it, for both
borrowed inputs and borrowed arrays.

* test(mcp): select the borrowed Evidence contract from its family

* docs(mcp): admit baseline growth only for restored contract declarations

* fix(mcp): state reduced multi-type properties as typed branches, not type lists

The reduced fallback derived a property's types and emitted them as a type
list, which Gemini, Moonshot and Ark reject. A property whose canonical
schema is anyOf string and number therefore traded a typed union for a
rejected construct when the inline projection fell back. Emit one type, or
anyOf typed branches, so [T, null] becomes the dialect's nullable union.

* feat(mcp)!: retire the full schema profiles and declare complete observations as a typed annotation (#10718)

* refactor(mcp): declare complete observations as a typed annotation applied only on the wire

* test(mcp): compare complete-observation declarations with their owning text

* refactor(schema): couple Zod checks with their JSON Schema constraint keywords (#10725)

* refactor(schema): couple Zod checks with their JSON Schema constraint keywords

* refactor(schema): declare pattern approximations and object output roots as typed annotations (#10754)

* refactor(schema): declare pattern approximations and object output roots as typed annotations

* fix(schema): refuse pattern declarations JSON Schema cannot state

* test(schema): select guided_workflows from its contract family

* feat(mcp)!: retire the full input and output schema profiles (#10841)

* feat(mcp)!: retire the full input and output schema profiles

The compact input presentation is now the only advertised input schema, and
output schemas are either omitted (`off`, the default) or bounded
(`compact`). Catalog identity, the build-recorded digests and the recorded
startup catalog collapse to those presentations: two digest profiles instead
of six, one recorded catalog instead of two, and no input dialect baseline.

Tests and verifiers that validated replies against listed full output
schemas now validate against the canonical output contract, which REA
already enforces server-side; tests that asserted canonical input
restrictions read the canonical input schema instead of a full listing.

BREAKING CHANGE: `REA_MCP_INPUT_SCHEMA_PROFILE=full` and
`REA_MCP_OUTPUT_SCHEMA_PROFILE=full` are refused at startup and by `rea setup`
with a configuration error naming the setting. Remove the setting from the
client registration (or rerun setup with `REA_MCP_INPUT_SCHEMA_PROFILE=compact`
/ `REA_MCP_OUTPUT_SCHEMA_PROFILE=compact`). `REA_MCP_INPUT_SCHEMA_PROFILE`
still accepts `compact`. Server-side validation is unchanged.

* test(mcp): validate replies against canonical output contracts at every listed-schema reader

tools/list omits output schemas by default, so fixtures, verifiers and the
Android analysis runners that read a listed outputSchema now bind the
canonical contract instead. connectLocalToolsMcp and advertisedMcpToolSchemas
hand the SDK a definition carrying the canonical output schema, so the SDK
client and caller-supplied validators check every successful reply.

* test(mcp): resolve Android runners' output contracts from the selected repository

The Android efficiency runners imported the canonical output-schema helper from the harness checkout while --repository selects the revision under test. Import it from the selected repository, as the runners already do for the SDK and Ajv.

* test(mcp): let Android runners compare revisions before the canonical output helper

A selected revision without scripts/lib/canonical-output-schema.mjs could not start the Android runners. Choose the output contracts in one harness helper: the selected revision's canonical contracts when it has them, otherwise its listed schemas under the full output profile it still supports.

* fix(scripts): keep caller launch settings when the real Hopper verifier builds its server environment

The verifier deleted REA_VERIFY_SERVER_COMMAND and REA_VERIFY_SERVER_ARGS_JSON from process.env itself before reading them, so a caller selecting an installed REA command silently ran the checkout's dist/main.js. Copy the environment, as main does, so only the child's environment changes.

* test(mcp): validate whole replies against canonical output contracts wherever a listed definition is held

tools/list omits output schemas by default and the full output profile is gone, so tests that handed the SDK a listed definition no longer checked complete replies. Attach the canonical output schema with withCanonicalOutputSchema at every held REA tool definition. Environment-served tools (not REA contracts), and the deferred-runtime and doctor tests that exercise per-profile listings, keep their listed definitions.

* test(mcp): record the advertised catalog summary without the retired full profiles

Retiring the full input and output profiles leaves only the compact input and compact output listings, so the summary drops the full-profile digests and every pattern only the full listings carried. Regenerated with npm run verify:advertised-schemas -- --update-summary.
2026-11-15 21:51:18 +08:00
morluto 3afd8a16ee feat(evidence)!: give Evidence bundles a format revision owned by one module (#10861)
* feat(evidence)!: give Evidence bundles a format revision owned by one module

`src/domain/evidenceFormat.ts` owns Evidence format evolution. Bundles now
carry `evidence_format: 1`; every bundle reader admits only the current
revision and refuses an unmarked or other-revision bundle once, with the
single recreate-observations guidance, instead of admitting earlier shapes
one optional field at a time.

The 6.7-era artifact inventory summary identities are no longer rebuilt:
compare_artifacts checks the current projection only and reports a mismatch
with the owner's guidance. The pre-6.6 identity diagnosis uses the same
guidance. SQLite `worker_timeout_ms` and Dart AOT `string_pool.policy`,
optional only for historical Evidence, are now required.

BREAKING CHANGE: Evidence bundles written by REA 6.x (no `evidence_format`)
are refused by import, comparison, reconstruction and analysis-snapshot
readers as an integrity mismatch telling the caller to recreate the
observations; re-exporting an earlier bundle does not update its format or
identities. Bundles constructed by hand must include `"evidence_format": 1`.
6.7-era artifact summaries are refused by compare_artifacts, and SQLite and
Dart AOT Evidence without the newly required fields fails validation.

* fix(evidence): let the ledger application admit its bundle's format

build_reconstruction_obligation_ledger registered the full bundle schema with the SDK, so an unmarked or other-revision bundle was refused as invalid_request before the format owner ran. Register it like coverage: the application parses the request envelope and admits the bundle.

* feat(pcap)!: read capture Evidence through the current result schema only (#10879)

Delete the pcap read-compatibility schema that defaulted executable identity for historical capture Evidence, and require query_labels_hex (null when unknown) now that every producer writes it. Evidence from an earlier format is refused at the bundle format owner; a record that omits these fields is reported as malformed instead of defaulted.

BREAKING CHANGE: inspect_pcap_capture results always carry query_labels_hex, and analysis views no longer default missing tshark/capinfos identity fields.

* feat(desktop)!: require the effective view policy in desktop state Evidence (#10886)

inspect_desktop_application_state always records the complete policy it used, including sqlite_view_bytes. Drop the optional historical shape and the task draft's absent-policy branch.

BREAKING CHANGE: desktop application state results without effective_view_policy, or with a policy missing sqlite_view_bytes, are refused as malformed.

* feat(android)!: require component declaration facets and reference aggregates in Android Evidence (#10904)

* feat(android)!: require component declaration facets and reference aggregates in Android Evidence

The JADX producers always write the manifest component projection with every declaration facet, target_definition and direct_invokes_coverage. Require them, delete the duplicate observed component schema, drop the unknown declaration coverage that only described historical omission, and give manifest view results their explicit facet.

BREAKING CHANGE: Android package, component and reference results missing these fields are refused as malformed; component declaration_coverage and its summary counts no longer include unknown; manifest view results always carry facet manifest.

* refactor(android): parse package results with the domain schema directly

* feat(browser)!: require the DOM length unit and redirect-scoped prior-hop facts (#10897)

* feat(browser)!: require the DOM length unit and redirect-scoped prior-hop facts

Both CDP producers always declare node_value_length_unit, and the observation session writes redirect_response on every redirect event and no other. Require the unit, drop the diff's absent-unit branch, and admit redirect_response exactly on redirect rows.

BREAKING CHANGE: DOM snapshots without node_value_length_unit and redirect timeline rows without redirect_response are refused as malformed, including inline inspections passed to compare_web_captures.

* docs(browser): drop the historical redirect_response omission note

* feat(native)!: require native Evidence fields that were optional only for historical Evidence (#10946)

Mach-O encryption, slice malformed cause, symbol runtime names and import
facts; ELF export encoding facts and visibility; value-flow limits; load-image
permissions and relocation symbols; Swift type-reference kind and GPU AMD
target metadata are required. Legacy enum values (unrecorded, unknown
visibility, undecodable, modinfo_field_exceeds_limit) and the retained
schemas, view defaults and reinterpretations that served historical Evidence
are deleted.

BREAKING CHANGE: native records missing these fields are refused as
malformed; recreate the observation.
2026-11-15 17:25:31 +08:00
morluto 8443930bd3 refactor(limits): declare the Node timer bound and duration schema once (#10640)
* refactor(limits): declare the Node timer bound and duration schema once

Fifteen modules re-declared 2_147_483_647 as their own timer maximum and
eleven built the same 1..2^31-1 millisecond schema. They now use
MAX_NODE_TIMER_DELAY_MS and timerDurationMsSchema from
src/domain/analysisResourceLimits.ts, and an architecture guard rejects the
literal outside that owner and five non-timer int32 bounds. The generated
MCP tool catalog is byte-identical.

* refactor(limits): keep the JVM processor-count bound separate from the timer bound

REA_JADX_ACTIVE_PROCESSOR_COUNT is a Java int count, not a timer delay, so it keeps its own named bound and a documented exception in the signed 32-bit guard. Remove the timer-bound and zod imports the duration-schema migration left unused.

* refactor(limits): keep the TShark packet-count bound separate from the timer bound

packet_limit is passed to TShark's -c as a C int packet count, not a timer delay, so it keeps its own named bound and a documented exception in the signed 32-bit guard.
2026-11-15 16:40:03 +08:00
Phonandmorluto 2fe0c9eb6a fix(cli): align local paths and SQLite selection validation (#9882)
fix(cli): batch of small fixes for #9714, #9718, #9759, #9856 and #9809

#9714: replace bare z.string().min(1) local paths in ADB CLI commands with localPathStringSchema so NUL bytes are rejected at the CLI boundary.
#9718: decode-android-resources path uses localPathStringSchema, matching apktoolInputSchemas.
#9759: inspect-sqlite-database CLI options now reject an empty column array and reject rowLimit/rowOffset/column/schemaScope=selected_table without --table via superRefine, mirroring inspectSqliteDatabaseInputSchema.
#9856 and #9809: shared snapshotSelectionOption is now localPathStringSchema.optional(), so every analyze/inspect/decompile/xrefs/trace/function/instructions/search --snapshot rejects NUL bytes.

Tests: tests/boundary/cli/overnightBatch1CliSchemas.test.ts (15 tests) proves empty/NUL input returns VALIDATION_ERROR before any provider runs and valid input is admitted.

Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-11-11 23:40:01 +08:00
morluto cb4e5cfbff feat(analysis): make resource budgets recoverable across CLI and MCP (#9706)
* refactor(analysis): centralize resource budgets and recovery facts

* feat(analysis): expose ASAR and Ghidra runtime admission controls

* feat(artifacts): expose reader admission and preserve limited recognition

* feat(javascript): expose recoverable semantic precision budgets

* feat(web): add recoverable source-map budgets and compact retained views

* feat(web): make historical capture decoding budgets recoverable

* fix(desktop): enforce selected view policies across capture and export

* feat(runtime): make Frida and LLDB retention budgets selectable

* feat(native): expose diagnostic budgets and preserve exhaustion causes

* feat(managed): expose recoverable snapshot and representation budgets

* feat(metadata): make Go and Rust inspection budgets recoverable

* feat(wasm): expose representation and producer admissions

* feat(evm): expose per-call worker resource budgets

* feat(android): make resource and string-pool omissions recoverable

* feat(producers): expose command and recorded-capture budgets

* docs(analysis): record recovery contracts and integrated validation

* fix(docs): link calibration oracle from published verification pages

* fix(docs): restore desktop task anchors and receipt links

* test(analysis): align fixtures with selectable budget contracts

* fix(analysis): remove source-location projector dependency cycle

* test(analysis): refresh budget recovery and verifier regressions

* test(analysis): preserve format and publication recovery expectations

* test(windows): align conformance with owned runtime admission

* fix(desktop): preserve selected alias kinds in exported state

* test(analysis): align provider fixtures with recoverable budgets

* test(desktop): validate emitted alias metadata structurally

* fix(ghidra): classify bounded Windows endpoint refusals

* test(cli): assert selectable artifact admission recovery

* fix(cli): preserve capture provenance in environment builds
2026-11-10 05:11:20 -05:00
3230f50b16 feat(sqlite)!: page selected rows with a stated order and row offset (#9331)
* feat(sqlite): page selected rows with a stated order and row offset

readRows read `NOT INDEXED LIMIT ?` with no ORDER BY and reported
order: "unspecified", so an agent could not reach rows past row_limit
deterministically: raising row_limit hits the fixed 16 MiB reply budget,
and the sampled order depends on whichever index SQLite happened to scan.

Read the window in a unique, stated order and expose the position:

- rowid tables order by rowid unless all of rowid, _rowid_ and oid are
  declared columns, in which case the order stays unspecified.
- WITHOUT ROWID tables order by the primary-key columns, reusing each
  key's DESC flag and declared collation from pragma_index_xinfo so the
  plan keeps index order instead of materialising a sorter.
- rows.order, rows.row_offset and rows.next_row_offset report the window;
  next_row_offset is null once the window reaches the end.
- row_offset (CLI --row-offset) requires table and defaults to 0. A
  non-zero offset is refused when the table declares no usable order,
  because the skipped prefix would be arbitrary.

The schema now checks next_row_offset against truncated and the offset so
a worker reply cannot claim a page it did not read.

One attribution fix came with it: adding ORDER BY text can push an
over-wide projection past SQLite's SQL-text limit, which fails as
SQLITE_TOOBIG before the result-column limit is reported. Attribute that
case to column capacity too, so a 250k-column request still reads as
capacity rather than a byte budget.

Verified: npm run check:fast, npm run format:check on the touched files,
npm run verify:module-boundaries, npm run docs:check, and 196 tests
across the sqlite boundary, application and CLI/MCP suites. A 1002-row
inspection and in-memory EXPLAIN QUERY PLAN checks confirmed the
WITHOUT ROWID path adds no sorter. Full check:pr and the complete test
suite were not run.

* fix(sqlite): resolve the rowid alias and keep paging claims reachable

Three defects in the row paging added alongside row_offset:

- readRowOrder emitted ORDER BY rowid for every rowid table, but a table
  that declares a rowid column shadows that spelling, so the sort used the
  stored column and could repeat or drop rows across pages. Order now names
  the first alias the table leaves unbound, read from raw name bytes so an
  unrelated malformed identifier is not decoded just to choose an order.

- The inventory projection could hide the primary key a WITHOUT ROWID table
  orders by, silently degrading a paginated read to unspecified order. The
  order key is now read from the table itself rather than from the columns
  the caller asked to see.

- A truncated table with no stated order reported a next_row_offset that the
  row_offset rule then refused. It is null there, and the reply limitation
  no longer claims an unspecified order for windows that have one.

rowEvaluationFailure also probed the first rows of the table instead of the
requested window, so a failure confined to a later page was attributed to
columns that evaluate fine there.

* test(sqlite): assert the concatenated pages

The paging regression checked only counts, offsets and flags, so a second
page that repeated the first page's rows would still pass. Assert the values
read across all three pages instead.

* fix(sqlite): decode row keys in the database encoding

readRowKeyFacts decoded primary-key names as UTF-8 and probed a fixed
hidden set, so two supported schemas could not be paged at all:

- A UTF-16 database stores key names as UTF-16 bytes, so "k" decoded to
  NUL-padded text and ORDER BY named an identifier that does not exist.
  Pass the detected encoding, as readColumns and the index-metadata path
  already do.
- A virtual generated column named rowid (hidden=2) was absent from the
  shadow check, so ORDER BY rowid evaluated the generated expression
  instead of the rowid and could sort non-uniquely or fail outright.
  Include generated columns in the alias probe.

Also stop decoding primary-key names for rowid tables: their ORDER BY
names an alias, never a declared key, so an unselected malformed key
name could abort an otherwise readable selection.

The page-concatenation assertion now narrows SqliteValue before reading
.value, which the null/blob arms do not carry; this was failing the
Static checks typecheck job.

* fix(sqlite): order WITHOUT ROWID rows by the index's own key sequence

`readRowOrder` filed `pragma_index_xinfo` key terms into a map keyed by
column name, so a valid composite key that repeats a column —
`PRIMARY KEY(a COLLATE BINARY, a COLLATE NOCASE DESC)` — collapsed to one
term. The generated clause then overstated uniqueness: it omitted the
BINARY term, so `A`/`a` ties survived, and the `row_offset` paging
advertised a `primary_key` order that the actual key does not establish.

Build the terms from the index rows in `seqno` order instead, keeping
repetitions, and bail to `unspecified` if any declared key column never
appears in the index. The declared collation stays explicit so a caller
column's own collation cannot re-sort the window.

Verified against a real SQLite database that the repeated key is accepted
and that the old clause orders A, a, B while the index orders A, B, a.

* fix(sqlite): accept a null next offset on unordered truncated rows

readRows already withholds next_row_offset when a table shadows rowid, _rowid_ and oid and declares no primary key, because every non-zero offset is refused there. The reply schema still required the offset whenever truncated was set, so the provider rejected its own worker result and the first sample never reached CLI or MCP. Require the offset only when the order is stated.

* fix(sqlite): probe row keys only for ordinary tables and name row_offset

readRowKeyFacts ran for a selected virtual shadow table before readRows
could reject it, so an unavailable module surfaced as SQLite's
"no such module" instead of the selection refusal. Probe the key only
when the selected kind is an ordinary table and leave the order
unspecified otherwise, since no rows are read for that kind.

The unordered-offset refusal also reported the generic selection reason,
which callers see against table. Give it its own reason so the failure is
attributed to row_offset.

* feat(sqlite)!: page selected rows with a stated order and row offset

Selected-row inspection currently reports only rows.order: "unspecified" and
offers no way to reach rows past the first sample deterministically. Main also
refuses non-zero row_offset today, so nothing relies on that omission; this
change makes the stated order and the offset the contract, reports the next
offset the caller feeds back, and keeps the existing byte, heap, statement and
deadline limits.

BREAKING CHANGE: rows.order becomes rowid | primary_key | unspecified, and
rows gains required row_offset and next_row_offset members. A client that
validates selected-table output against the previous schema (where order was
the only permitted value) rejects a now-valid reply. Migration: accept the two
new members and treat order: "rowid" | "primary_key" as the stated order the
offset applies to; a null next_row_offset means the window reached the end.

* docs(sqlite): say a null next offset also means no stated order

The selected-rows limitation described next_row_offset as null only when the window reached the end. An unordered table whose row_limit truncates the sample also publishes null, because an offset there would name a request the row-offset rule refuses, so the Evidence read as if the sample were complete. State the second cause and pin it in the existing regression.

* test(sqlite): drop the limitation wording assertion and state the unordered exception

The boundary test pinned the sentence a caller reads in `limitations`,
so a harmless rewrite of that text failed a paging test whose real
subject is `order`, `truncated` and `next_row_offset`.
docs/development/testing.md asks for facts over frozen explanatory prose,
and the three structured expectations already fail if the paging fact
changes.

The tool contract promised that retained rows always have a unique order
and can be paged. A rowid table that shadows rowid, _rowid_ and oid and
declares no primary key has no such order: it returns
order=unspecified with a null next offset and refuses a non-zero
row_offset. Say so in the description so callers do not plan a
continuation the tool always refuses.

* docs(sqlite): say unordered tables cannot be paged

The description told callers that a truncated table without a stated order
needs "the table read without offsets", but omitting row_offset starts at
zero again: with an unchanged row_limit the next call repeats the first
sample instead of continuing past it. Say that these tables cannot be
paged and point at row_limit for a larger first sample.

* style(cli): wrap the native data type path description

The `inspect-native-data-type` argument description line is 82 columns, over the repository's 80-column `.oxfmtrc.json` limit, so `npm run format:check` fails on main and on every branch merged from it. It came in with #9465 and the following commits to this file did not touch it. Reformat the one call the formatter objects to; no behaviour change.

* fix(sqlite): bound rowid alias comparison by name byte length

readRowKeyFacts decoded every column name before testing it against three short aliases, so a table declaring a multi-megabyte name could exhaust the worker heap even when the caller projected the column away. Reject names that cannot be an alias by their byte length, accounting for the database encoding, before building any string.

---------

Co-authored-by: 2SinceJuly <202121119+2SinceJuly@users.noreply.github.com>
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
2026-11-09 21:01:58 +08:00
gensokyo-dev e18594b247 refactor(composition): unify capability bindings and cleanup ownership (#8441)
Bind typed service creation, observed availability and retryable cleanup in one composition owner. Reuse selected CLI/MCP factories and canonical contract groups without copying provider operation inventories.

Closes #8429. Refs #740.

Validated paired real application workflows, actual local Codex CLI Terra/high use, complete Evidence and Unknown checkpoint parity, compiler-backed import boundaries, SDK schemas and final real mixed routing. Automated review findings were fixed at their producing owners. Existing exact-base CodeQL alerts are documented on the PR under the owner-authorized unrelated-check exception; no global alerts or security rules were changed.

Assisted-by: Suika (gensokyo-skills:suika-scatter-gather-planning)
2026-11-01 13:30:21 +08:00
morluto bd4f5e9c19 fix(parsers): preserve format evidence and caller input identity
Retain bounded FAT and offset-tagged PE observations through archive hashing,
classification, and content nodes. Preserve PyInstaller member paths according
to the observed carrier dialect and share APK v1 signing-name semantics.

Read SQLite's implicit sqlite_schema table without inventing a stored schema
declaration. Distinguish unsupported ELF numbering and decoded-name admission
from malformed data and fixed scan allowances. Preserve Sigrok basenames and
classify matching explicit input-format refusals at their owning boundary.

Validation: 554 distinct tests passed, 3 skipped across serial focused runs;
build, typecheck/lint, docs, formatting, and discovery (2141 files) passed.
Sigrok command-boundary simulators do not establish real-tool decoding.
Small repeated public PyInstaller CLI runs had equivalent inventory facts;
warm median elapsed 0.5431s before / 0.5426s after, with no speed claim.

Fixes #6733
Fixes #6741
Fixes #6722
Fixes #6732
Fixes #6736
Fixes #6734
Fixes #6764
Fixes #6766
Refs #6739
2026-10-23 22:54:28 +08:00
郭立 Guo Li 7590b50801 perf(schema): reduce retained integer validator allocations (#4360)
Construct integer formats directly while preserving bounds, refinements, metadata and current shared schema owners. Keep Incur-dependent CLI numeric coercion declarations unchanged.

Verify canonical advertisements, 126 numeric boundary cases, 974 actual CLI parser cases, broad source tests and real public-artifact SDK workflows. Record 72 normal tasks and 12 separate owned-inspector heap diagnostics with whole-family resource bounds; distinguish SDK spans and asymmetric audit overhead. Generate numerical report sections from receipts and reject stale tables through documentation conformance checks.

Closes #6236. Current-head CI and Greptile pass; actionable review findings are addressed. Codex review was unavailable due repository usage limits.
2026-10-22 05:11:08 +00:00
morluto cdd718ffb0 refactor(inputs): own absolute path validation in shared contracts 2026-10-22 08:13:46 +08:00
gensokyo-dev f801b2fbfd feat(sqlite): make worker deadlines adjustable with recovery guidance (#5818)
Expose a per-call SQLite worker deadline with exact CLI/MCP recovery controls,
parent-owned execution metadata and historical unknown preservation. Keep fixed
resource limits and cleanup admission intact. Share typed cleanup composition
with one captured-output owner and preserve the timeout after cleanup resolves.

Fixes #5812. Follow-up tracking: #5808. Existing CLI test-oracle failure: #5845.

Verified 155 SQLite tests (3 skipped), static/docs/discovery checks, six real
CLI/MCP pairs and 13 actual Codex sessions including four corrected deadline
recoveries. Three pre-dispatch discovery failures remain explicit, not passes.
Review findings include real retained-cleanup and fixture ownership regressions.

Assisted-by: Kogasa (gensokyo-skills:kogasa-surprise-testing)
2026-10-21 16:45:46 +08:00
morluto 204b9f3ac1 fix(contracts): preserve array guidance across MCP schema profiles
Keep item schemas when compact inputs reduce nested structure, expose literal tuple elements to items-first renderers, and preserve SQLite table-selection guidance across compaction. Verify catalog-wide client constraints and canonical rejection before dispatch. Document the stock Pi renderer size limit and caller-selected recovery.
2026-10-15 01:41:31 +08:00
YsopwnableandY0UW1SH aeee5074d1 feat(sqlite): inspect offline snapshots with selective evidence (#2271)
Add offline SQLite database/WAL inspection through the shared CLI/MCP.

Use compact inventory, selected-table schema, exact ordered columns and
bounded row samples while preserving source identity and lossless values.
Retain completed Evidence through late cancellation and private-root cleanup
failures, bound concurrent inspections, and recheck journal validity after
acquisition. Reuse prepared CI runtime artifacts across supported Node lines.

Co-authored-by: Y0UW1SH <29498139+Y0UW1SH@users.noreply.github.com>
Assisted-by: Kogasa (gensokyo-skills:kogasa-surprise-testing)
2026-10-12 18:33:23 +00:00