The vote suite grew from 375 tests (39.1 min on one runner, 2026-09-30)
to 1,321, so a single-runner vote can no longer finish. Run it as the
same ten duration-planned shards playwright-mock.yml uses; each shard
logs the specs it executed and the codegraph shadow unions them into
one trial. A shard that runs out of its 28-minute test budget logs only
its failed specs, never clean ones.
* 💉 chore: Upgrade Node.js to Security Release 24.22.1
* test: Enable Enter Submission in Native Acceptance
* test: Use the Machine Picker in Native Acceptance
---------
Co-authored-by: Lia <lia@librechat.ai>
Each merge to dev cancelled the vote run before it, and the full mock
suite needs about 43 minutes, so 480 of 499 runs since 2026-09-24 never
finished (282 superseded, 153 cut by the 45-minute timeout, 45 never
started). Graduation evidence stopped at 09-30. Keep a started run and
one pending run, and raise the timeout to 75 minutes.
The step appliers, frame normalizer and SSE transport move from client/src/hooks/SSE into packages/chat/src/core with their specs, and the client imports them from @librechat/chat. The ask-user-question predicates the content applier reads move with them, and the client's approval module re-exports them so its consumers are unchanged.
* feat: Scaffold the @librechat/chat package
Adds packages/chat with core, react and components entries, a tsdown build,
a node-environment jest config, turbo and root build wiring, import boundary
rules, design-lint scope, and core types for the chat transport, stream,
host and status. No code moves yet.
* test: Cover the @librechat/chat boundaries as scenarios
* fix: Run the chat suite in test:turbo and build it for the mock lane
test:turbo now selects @librechat/chat, the mock e2e job builds the package the
chat-package scenarios load, and the synthetic repository in the suppression
scenarios carries the packages/chat/src root lint:design:record names.
* chore: Build the chat package in frontend:ci too
* chore: Run the chat suite in test:all
* fix: Hold core TSX to the chat boundary and scan the package for cycles
* fix: Apply the chat import boundary to JavaScript sources
* fix: Keep nested core files off the components entry
* fix: Close the remaining chat package gates and fence aborts by generation
Blocks imports through the @librechat/frontend workspace name, adds the chat
sources to the i18n scan, the CSS color gate and the client Tailwind content,
and lets ChatAbortRequest carry the generation it means to stop.
* fix: Keep JSX out of the chat core and carry abort and reconnect outcomes
Rejects JSX under the core, copies the chat manifest before the Docker
install, keys the client app build caches on the chat sources, fences
reconnects by generation, and types the abort result the transport returns.
* test: Give the CSS gate's scratch tree the chat root it now scans
* fix: Close the remaining @librechat/chat gates and wiring
Rejects import() and require() across the package boundaries, holds chat
components to the literal-copy rule, compiles TSX in the package's jest
transform, builds the package in Frontend Unit Tests, requires its test job for
main promotion, adds it to the unused-dependency scan, and lists it wherever a
sibling workspace is listed: Dockerfile.multi, the docker smoke paths, the
remaining node_modules caches, the reinstall helper, the bun build chain and
AGENTS.md.
* fix: Tighten the chat boundaries and carry steering and Assistants aborts
Template-literal import() and require() and self-imports through the package
name are rejected, user-facing JSX attributes need localized text, the cycle
scan resolves self-imports, Dockerfile.multi builds the package for the client
stage, and the transport carries steer, cancelSteer, armSteer and the
Assistants abort key.
* feat: Match the chat transport and host contracts to the client's
The transport now reports settled and replaced starts, the stream's protocol
version and resumed flag, reconnect replay, generation status, paused-run
resume with approvals or answers, abort and steer protocol versions, and the
request error shape. The host config and auth carry what the client's
ChatSettings now supplies: rename-while-running, model specs, retention mode
and the user.
* fix: Keep lazy React-entry loads out of the core and type the Assistants stop
Relative import() and require() of the /react entry from core are rejected,
and an abortKey stop resolves to the terminal conversation and run messages the
Assistants route returns.
* fix: Fence start controls by protocol and close extension-bearing entry loads
Settled and replaced start results carry the protocol version, core loads of
the /react or /components entry with a file extension are rejected, and
test:packages:chat builds data-provider first.
* 🖼️ fix: Precache PWA Icons and Let Them Refresh
Production builds set `publicDir: false`, so `client/scripts/post-build.cjs`
copied `public/assets` into `dist/assets` only after `vite build` had finished.
vite-plugin-pwa globs `dist/` from its `closeBundle` hook, inside that build, so
the icon patterns in `workbox.globPatterns` matched zero files and the generated
`sw.js` shipped without a single icon, silently: workbox treats an empty glob as
a no-op. `build:ci`, used by the Playwright lanes, never ran the post-build step
at all and produced a `dist` with no icons or `robots.txt`.
Move the copy into a `copyPublicAssets()` build plugin that runs in
`writeBundle`, which completes for every plugin before `closeBundle`, and drop
the post-build script with its lone `fs-extra` dependency.
Ordering alone was not enough. The plugin defaults `dontCacheBustURLsMatching`
to `/^assets\//`, true only for hashed bundle output. The icons live in
`assets/` under stable filenames, so they entered the manifest with
`revision: null`, permanently immutable, meaning an installed PWA would never
pick up a rebranded icon: the other half of the report. Match Vite's
`[name].[hash].[ext]` shape instead so hashed chunks stay revision-free while
the icons carry content revisions.
Also drop `manifest.webmanifest` from `globPatterns`, since the plugin always
appends it as an additional manifest entry and globbing it duplicated the
precache entry.
A clean `npm run build` now yields 55 precache entries with all five icons
revisioned, against 0 icon entries before.
* 📝 docs: Describe the Full Copy Set in copyPublicAssets
* 🔒 chore: Drop fs-extra From bun.lock
First live full-suite vote run (32731087273, 198 passed) wrote
e2e/pw-results.json while the ledger step looked in the repo root and took
the fail-safe branch: zero trials logged. A relative
PLAYWRIGHT_JSON_OUTPUT_NAME resolves against the config directory, not cwd —
reproduced synthetically with the config in a subdirectory. Absolute
workspace path on both the reporter env and the ledger read.
* ci: votes run the full mock suite; covered list from Playwright's own discovery
The vote workflow passed the merged PR's skippable tier as CLI path filters, but
playwright.config.mock.ts scopes discovery to testDir specs/mock/ — tier entries
outside that directory matched nothing, and the covered-list log line still
claimed them. Run 32701691037 proves it: a11y/keys/messages in the covered list,
zero of their tests executed, '120 passed' all from specs/mock/. The graduation
ledger was minting clean trials for specs that never ran.
Now every dev push runs the full mock suite (no path filters to mismatch), the
covered list is derived from playwright --list --reporter=json (git-enumeration
fallback over the same testDir), and each merge is one trial for every pool spec
— ~4x faster accrual toward the pre-registered graduation bars, plus the
post-merge Playwright safety net the jest workflows already have. Timeout 30->45
for the wider run; newest merge still cancels older votes; observe-only,
continue-on-error, kill switch CODEGRAPH_E2E_VOTES unchanged.
* ci: covered list from executed results, not discovery (Codex P1)
Env-gated suites (mcp-tool-list-changed needs E2E_MCP_LIST_CHANGED, enforced-
model-specs needs E2E_MODEL_SPECS_ENFORCE) are discovered by --list yet skip
every test under the vote job's default env — counting them as covered would
mint phantom trials, the exact class this PR exists to kill. The run now emits
line+json reporters and the ledger step derives covered from specs with at
least one non-skipped test outcome; no results json means no trials logged.
Verified against a synthetic suite: gated spec excluded, nested dirs handled,
crash branch logs nothing.
Every Playwright job spent a flat 90s on `npx playwright install ffmpeg`,
and none of them ended up with a usable ffmpeg.
The 2.3MB download finishes in under a second; extraction then hangs until
`timeout -k 10 90` reaps it (exit 124, masked by `continue-on-error`). That
is a Node 24.16.0 readable-stream change (nodejs/node#62557) colliding with
yauzl/fd-slicer never firing `close` after EOF, which hangs extract-zip.
It leaves a truncated `ffmpeg-linux` — 5,055,201 bytes against the zip's
declared 5,101,056, segfaulting on exec — and no INSTALLATION_COMPLETE
marker, so Playwright treated ffmpeg as uninstalled. `video: 'on-first-retry'`
has therefore never worked in CI, and every first retry of a flaky test died
in browserContext.newPage: exactly the failure the step existed to prevent.
Upstream fixed it in Playwright 1.60.0 (microsoft/playwright#40747) and Node
reverted it in 24.18.0 (nodejs/node#63834). Node 24.16.0 is pinned in 17
places including the Dockerfiles, so bump Playwright instead — it is a dev
dependency, and `^1.56.1` already permitted 1.62.1; only the lockfile pinned
it. Staying at or above 1.62.1 also avoids the tsconfig-resolution
regressions in 1.62.0.
Caching alone could not have fixed this: a cold cache still hangs, and what
would have been cached is the corrupt binary. So the ffmpeg download is now
restored from cache keyed on the resolved playwright-core version, the
install is skipped outright on a hit, and the cache is only saved once the
binary is verified to actually execute — a partial extraction can never be
promoted into a cache that every later job restores.
Per job: 90s to ~0s on a hit, ~2s on a miss.