* fix(javascript): follow bundler namespace interop for Electron APIs
Rolldown/tsdown emit `let electron = require("electron"); electron =
__toESM(electron);` and esbuild/tsc emit `__toESM(require("electron"))`
or `__importStar(require("electron"))` for namespace imports. The
reassignment or the helper call hid the Electron origin, so bundled
main processes reported zero BrowserWindows and preloads.
Locally declared, never reassigned interop helpers recognized by name
now pass the module through, and a binding's rewrap of its own value is
not treated as a reassignment. Other writes stay unresolved.
Fixes#1802
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(javascript): require emitted interop helper definitions
Only trust an interop helper name when it has a single program-scope
function definition, and state that helper bodies are not evaluated.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(bun): report the real runtime and enforce worker memory budgets on Windows
Under Bun, REA recorded Node/V8 identity it did not observe and its JavaScript helper processes ran without a memory bound, because Bun ignores the V8 heap flags REA passes. Source-map trace Evidence claimed runtime node-v8 with Bun's synthetic V8 heap limit (the decoder then rejected that limit, so trace_web_source_location always failed), analysis resource failures reported heap limits Bun does not enforce, the HAR decoder claimed a fixed 192 MiB heap, and doctor reported Bun's Node.js compatibility version as the Node version.
Identify the runtime once from process.versions. Source-map reports record id bun with the Bun and Node.js compatibility versions and a null v8_heap_limit_bytes; the decoder checks the heap budget only for V8 reports and rejects a report from a runtime other than the launching one. The analysis worker reports a null heap limit under Bun, HAR limitations and failures omit heap sizes under Bun, and the doctor node check reports Bun and passes when its Node.js compatibility version is supported.
On Windows under Bun, each helper's owned Job Object now carries a commit limit of its heap budget plus a measured 512 MiB Bun runtime reserve (704 MiB for the codec and HAR decoder, 1536 MiB for the default analyzer). The native addon accepts an optional page-aligned job memory limit on process_spawn, verifies it by readback, associates a private completion port before the child is created, and reports a sticky memoryLimitReached and peak job commit on each poll; inspect reports jobMemoryLimit. A requested limit with an artifact lacking the control refuses the launch, and non-Windows launchers refuse a limit.
Reaching the limit is classified by the job notification rather than Bun's exit status: the analyzer fails the current source with a memory resource failure reporting worker_process_memory_limit_bytes, the codec returns a memory resource failure, and the HAR decoder returns its memory failure with process_memory_limit_bytes. Bun source-map Evidence records process_memory_limit_bytes, verified against the launcher. Node.js launches and outputs are unchanged.
* fix(windows): drain memory notifications after observing exit
---------
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
* feat(javascript): recover code-split bundles from multiple inputs
recover_javascript_sources accepted a single path, and
validateModuleSet bound every module's provenance to that one snapshot.
A code-split app (runtime + chunks, or an Electron main/preload/renderer
set) had to be recovered one file per call, and modules recovered alone
could not resolve imports across chunks: cross-chunk require(id)
references stayed as numeric ids with property reads.
Add an optional additional_paths input. The listed files are
concatenated, in order after path, into the single engine snapshot, so
one Wakaru run sees every module and cross-chunk imports resolve.
validateModuleSet's single-snapshot binding is unchanged. Module
provenance is attributed back to the input file its first extraction
range starts in, with byte ranges rebased to that file's own offsets;
a module whose ranges cross an input boundary keeps the concatenated
snapshot identity and offsets, recorded as an explicit limitation.
Single-path behavior is byte-for-byte unchanged (one segment,
offset 0).
Fixes#1605
* fix(javascript): preserve source identity in multi-input recovery
---------
Co-authored-by: JingHao-Leon <w15909130000@gmail.com>
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
Review follow-up. Only a capture already on the alias chain can repeat a prefix, so a first visit no longer slices and serializes every suffix of its path.
Detect unprunable repeated self-alias prefixes at the JavaScript mutation owner, record conservative escape effects, and reuse settled repeated-prefix checks. Preserve finite-path precision and deeper-write uncertainty.
Refs #1665.
Co-authored-by: Martin Klein <36695017+flitzrrr@users.noreply.github.com>
* fix(build): emit no runtime imports for type-only bindings
Under verbatimModuleSyntax, an import whose specifiers all carry inline
`type` modifiers compiles to `import {} from "..."`, a side-effect import
that still resolves and evaluates the module. 78 compiled modules carried
one, including `import {} from "playwright-core"` in the browser scenario
session, which loaded Playwright's module graph at MCP startup.
Use top-level `import type` for these bindings and enable oxlint's
typescript/no-import-type-side-effects rule so the pattern cannot return.
Refs #1685
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* perf(mcp): load Playwright when a scenario launches, not at startup
The browser scenario and Electron active providers are constructed before
the stdio server serves, and both bound Playwright's launchers at module
load. Resolve `chromium` and `_electron` from playwright-core when a
capture opens instead, as the web module resolver already does.
MCP startup now resolves 1,312 modules (4,902 resolutions) instead of
1,598 (6,076) before reading the first request, and no longer loads
playwright-core. Provider construction, the tool catalog, and capability
inventory are unchanged.
A boundary test spawns the stdio server, initializes, lists tools, and
asserts playwright-core and isomorphic-git were not resolved; the CLI
probe now exercises the real launch path to prove detection.
Closes#1685
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
`import electron from "electron"` bound the module with a leading `default`
export segment, so `electron.ipcMain.handle(...)`, `new
electron.BrowserWindow(...)` and destructuring from that binding matched no
Electron API. Electron's CommonJS exports define no `default` member, so the
segment can only denote the module object; drop it when classifying
Electron bindings.
The binding classifier already recorded optional-call roots, but the IPC,
context-bridge, utility-process and sender-validation visitors accepted only
plain calls. `ipcMain?.handle(...)`, `contextBridge?.exposeInMainWorld(...)`
and the common `event.senderFrame?.url.startsWith(...)` check were dropped
without a limitation.
Refs #1656
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* feat(javascript): recognize tRPC procedures and GraphQL operations
Record RPC client syntax as network endpoint candidates in both
analyze_javascript_application and analyze_web_bundle, distinguished by
mechanism:
- trpc:procedure:<method> keeps the static member path of
root.<path>.<method>(...), which tRPC proxies preserve through
identifier minification. React hooks and TanStack option factories are
recognized directly. Vanilla query/mutate need a root bound to
createTRPCClient/createTRPCProxyClient, or a tRPC source marker with a
router path of at least two segments; subscribe needs the marker and a
tRPC observer argument. `this` roots, browser globals and RTK Query
`endpoints` chains are never procedures; cache-key helpers are excluded.
- trpc🔗<link> records the literal url of documented HTTP links.
- graphql:<type> records named operations from string and cooked template
literals with a valid operation header, and from precompiled
OperationDefinition objects; a graphql-tag loc.source.body copy is not
reported twice.
Method, callback and factory names follow @trpc/client and React/TanStack
integration sources and documentation for 11.19. Carrier segments before
the router path, useUtils helpers and minifier-renamed links remain
documented limitations.
Validation: typecheck; domain tests for recognition and false positives
(Apollo, RxJS, stores, RTK Query, browser globals, prose); web bundle
analyzer test; end-to-end analysis of a minified fixture through the
built MCP server on Windows 11.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* fix(javascript): preserve RPC binding identity and bundled calls
* style(reference-source): fix CI formatting
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
Reuse lexical ownership to exclude shadowed require calls while retaining native CommonJS and direct createRequire bindings. Preserve candidate edges when JavaScript or TypeScript runtime mode is unknown.
Co-authored-by: Rudy Celekli <rudy@gradiahq.com>
Resolve values returned by simple assignment chains at the export boundary. Suppress property links proven to target a detached receiver; retain unresolved links when reassignment may preserve receiver identity. Identify root assignments from their AST target rather than the displayed export name.
Validation: 25 semantic tests, seven targeted compiled CLI/MCP scenarios with Node runtime controls, build, and check:fast passed. Existing full acceptance suite passed before receiver refinements; the final changes were checked with targeted scenarios rather than repeating its 130-second run.
Resolve this, private fields and super reads through receiver-aware references. Separate instance fields from prototype descriptors, preserve alternative inheritance branches and allow finite repeated getter paths.
Add domain precision regressions and compiled CLI/MCP coverage for the four reported getter shapes.
Fixes#1665
Reuse semantic module provenance for aliases and keep property path segments intact through Electron matching. Reject shadowed names and nested or dotted lookalikes. Reassigned and conditionally initialized module origins remain ambiguous, with their source origins retained.
Add a compiled CLI/stdio MCP regression for windows, preloads, context bridges, IPC operations, pairing, and source ranges. Validate 771 JavaScript domain tests, seven composition tests, the public acceptance workflow, build, and check:fast.
Fixes#1656
Replacing or deleting a class getter property does not mutate the object it would return. Apply the existing object-accessor write boundary to instance and static getters.
Validation: source mutation regressions and compiled CLI/stdio MCP acceptance preserve literal 1 for all four assignment/deletion cases; check:fast passes.
Track active bindings with per-evaluation push/pop instead of copying ancestor sets on every alias. Restore membership in finally so cycles and sibling branches keep their existing semantics.
Validation: 173 domain tests, compiled CLI/stdio MCP returned-reference acceptance, and check:fast pass. A randomized six-block Flameox experiment retained identical complete semantic output and reduced the bounded 112-alias analyzer median from 36.33 ms to 17.00 ms.
Follow object, instance, and static getter results when invalidating shared values. Respect descriptor replacement, known spread overrides, inheritance alternatives, and unresolved computed keys. Preserve conservative uncertainty for fresh getter results and verify shared mutations through CLI and MCP.
Fixes#1494
Assigning a new value to the exports alias does not replace module.exports. Stop inventing a default export link and definite return-shape changes for that assignment.
Validation: 22 semantic structure source tests and 49 compiled CLI/MCP acceptance tests passed. The new regression checks Node require returns an empty export object for both versions and both adapters report missing defaults without definite changes.
Build each side's digest-to-Evidence index once instead of scanning native observations for every comparison item. Retain every exact match and avoid linking subjectless Evidence to nodes lacking content identity.
Comparison and relationship regressions pass. Isolated alternating unprofiled runs at 2,000 nodes and observations per side have conventional medians 379.924 ms before and 246.466 ms after; the complete 4,000-item outputs match byte-for-byte. Timings are bounded workload evidence, not whole-suite performance.
Flameox profile: flameox://evidence/7dca714833dc3d254a75704670a81de656b4b31d4b77a72236eecb82459b3d86
Avoid repeatedly canonicalizing scalar seeds for every literal node. Valid literal metadata without a scalar value now remains unmatched instead of throwing.
Flameox identified canonicalization as the hotspot. Alternating warm query benchmarks over 2,002 nodes reduced the 35.5 KB seed median from 36.34 ms to 0.158 ms; a short seed improved from 0.304 ms to 0.128 ms with identical result digests. These measurements cover the query function, not end-to-end analysis.
Validation: 38 domain query/traversal tests and focused compiled CLI/MCP consumers; static, architecture and documentation checks passed.
Use resolved seeds directly, retain one list of relevant unknowns, and stop sorting adjacency used only for complete reachability. Preserve canonical graph-array output order and candidate coverage scans.
Validation: 37 query and traversal tests, public JavaScript reconstruction and invalid-Evidence consumers, check:fast. No real-browser or deep-provider workflow verified.
Scan eligible candidate relations directly for query ambiguity instead of allocating and sorting another traversal index. Preserve direction, relation filters and cycle handling.
Semantic graph and builder suites plus public application workflow CLI tests passed. Full query outputs matched on a generated 20000-node graph; the single repeated-query timing run is indicative only.
Construct typed semantic records directly from cloned, normalized input and derived commitments instead of reparsing the same fields. Keep full graph integrity validation and verify caller mutation isolation. Four alternating cold runs preserved complete result digests while median projection time fell about 8 percent and 17 percent for the two 20,000-node capacity workloads.
Append overlapping observations to node buckets instead of recopying groups, and share one node index across semantic seed admission and traversal. Preserve merged provenance and absent-seed behavior.
Detach cancelled CDP sessions within the cleanup budget and retain event-finalization failures across retries. Honor Git filesystem encodings and exclude the promise-API probe from source-read failures.
Align existing integration regressions with retained process resources, filesystem-order-independent truncation, owned skill files, explicit native access facts and streamed JSON errors.
Validation: 132 acceptance tests passed (2 skipped), 83 affected boundary tests passed, Git/JSON/CLI boundary regressions passed, check:ci, docs:check and test discovery passed. Local verification used Node 24.18.0 on macOS; cross-platform CI remains pending.
* fix(electron): resolve pages served from inside app.asar
Packaged Electron applications load their renderer pages from file URLs
inside app.asar. The host filesystem cannot stat those paths, so
list_electron_targets excluded every such page as unsupported_url and
inspect_electron_page could not select it.
Resolve these URLs the way Electron's file protocol does: the first .asar
path component that is a regular file is the archive, and the remainder
must name a regular file in its header. Directories, links, and missing
members stay rejected. Parsed headers are cached by archive identity.
* fix(artifacts): extract and explain rewritten ASAR unpacked companions
Packaging tools commonly sign or strip .asar.unpacked native binaries after
writing the archive header, so their declared hashes no longer match.
analyze_javascript_application already accepts integrity_policy, but
extract_artifact did not, and inspect_artifact and extract_artifact still
told callers to "get a fresh copy", which cannot help for a shipped app.
- Accept integrity_policy on extract_artifact (MCP and CLI). The default
still fails closed; record-and-continue writes the observed bytes and
returns the extracted files' integrity_contradictions records.
- Name the unpacked entry and the post-packaging cause in integrity
failures, and point every operation that accepts integrity_policy at
record-and-continue.
- Report an active unpacked entry whose companion file is absent as
unavailable with its logical path instead of a format failure.
- Share one integrity_policy schema across the inventory, extraction, and
JavaScript application inputs.
* fix(electron): bound ASAR headers before allocation
---------
Co-authored-by: morluto <76467478+morluto@users.noreply.github.com>
HTML audio and video elements have their own `preload` attribute ("none",
"metadata", "auto"). A React renderer compiles `<audio preload="metadata">`
to `jsx("audio", { preload: "metadata" })`, and the `property:preload`
heuristic reported it as an Electron preload script named `metadata`.
Skip `preload` properties in the props object of a call whose first
argument is the static tag `audio` or `video`. The callee is not checked,
because minified element factories have arbitrary names.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01XPY2GmufMNGQKoEgRUx1pd
After #1542 the longest event-loop blocks on large bundles were single AST
traversals inside one analyzer: member-mutation collection with its
fixpoint, static source analysis, and function fingerprints, each about
one second on Obsidian's 3.7 MB app.js.
- traverseJavaScriptAstSteps yields every 1,024 visited nodes with the
same visit order; traverseJavaScriptAst drains it, so every existing
caller is unchanged.
- Member mutations, static analysis and fingerprints run as steps. The
member-mutation fixpoint yields every 64 references, and fingerprints
yield between callables.
- The now-unused synchronous member-mutation and fingerprint wrappers are
removed.
Graph IDs and complete result digests match main on three inputs,
including obsidian.asar. Over two runs there, the longest block falls
from about 920 ms to 740 ms, the second from 860 to 595 ms, and blocks
over 250 ms from 27 to 18, with total time within 1.5%.
Refs #1462
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix(javascript): yield inside semantic analysis and large projection passes
After #1525 the longest event-loop blocks were single stages on very large
files: semantic analysis of Obsidian's 3.7 MB app.js ran 5.8 s in one piece,
and its object projection 1.8 s.
- analyzeParsedJavaScriptSemanticsSteps yields between the analysis phases
and between the eight derived collectors; the synchronous function drains
it, so results are unchanged.
- Per-file analysis yields between open-receiver classification and static
analysis and drives the semantic steps.
- Semantic node creation, definitions and references, and the object and
frontier passes yield every 64 items. Node creation keeps source order,
which the per-file node budget depends on.
On obsidian.asar the longest block falls from 5,355 ms to 907 ms with the
same total time (40.6 s to 39.5 s). Graph IDs and complete result digests
match main on five inputs. Remaining blocks up to about 0.9 s are single
AST traversal passes inside the analyzers.
Refs #1462
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
* refactor(javascript): drop the unused synchronous derived-semantics wrapper
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
---------
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
analyze_javascript_application ran several long phases with no event-loop
turn, so MCP ping and cancellation waited: each file's parse, analysis and
semantic projection; building every application node and relationship
before the application graph's first sealing step; and whole-graph schema
parses inside both immutable graph factories.
- Each file runs as steps (parse, static analysis, semantic analysis, then
each projection pass), driven by the existing time-sliced checkpoint.
- Application graph input construction yields between phases, per module
link and per file of findings.
- The pipeline transfers its privately built graph inputs to new sealing
steps that validate the graph's own fields synchronously and elements,
content checks and the digest in steps. The public copying factories
keep their synchronous capture.
Results are unchanged: graph IDs and the complete result digest match main
on five inputs including Obsidian's 23 MB obsidian.asar. On the issue's
1,200-function fixture the longest event-loop block falls from about
290 ms to 90 ms, and ping after abort from up to 286 ms to at most 56 ms.
On obsidian.asar it falls from 10.9 s to 5.4 s; the rest is one analyzer
stage of a single very large file.
Refs #1462
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: N0zoM1z0 <161784452+N0zoM1z0@users.noreply.github.com>
Since #1471, escaping an object also escapes every method's results. Each
call through the object escapes it as the receiver, and each escape
re-expanded all of its methods. N methods returning one shared object,
each called once, did N times N work: 2,000 calls took 3.0 s of semantic
analysis and 4,000 took 12 s.
Expansion below references depends only on syntax, so escapes now reuse
it per expression and path. Identifiers that resolve to the same binding
and path collapse to one, because the escape walk treats them alike.
Every escape site is still recorded. The semantic IR is byte-identical
on 50 sources, and the issue's case takes 253 ms at 2,000 and 638 ms at
4,000.
Closes#1495
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
* fix: bind identity and coverage to the producing rule
Logical paths, container families, examined bytes, path-segment conventions, destination collisions, file-id leases, Mach-O CPU selection, and firmware cleanup now follow the observation that owns each rule.
* fix: fail unmatched Mach-O slices and keep firmware roots independent
A fat target whose architecture lipo did not list now stops before slice tools run. Chained fixups still count unread LC_DYLD_INFO rebase bytes, and a firmware cleanup failure no longer blocks the next workspace.
* test: cover path identity and runtime families at the projection boundary
NFC joins and segment conventions are now asserted by the path resolver, the Apple projection, and a real APK inventory instead of helper unit tests.
* test: expect unread dyld rebase bytes in the Apple dispatch fixture
The legacy LC_DYLD_INFO fixture keeps a rebase range this reader does not decode, so the public CLI and MCP check now requires that partial coverage beside the bind opcodes.
---------
Co-authored-by: Cursor Agent <cursoragent@cursor.com>
