Files
gstack/.github/workflows/evals-periodic.yml
T
01593aa67c v1.91.2.0 fix: consolidate gstack reliability wave (#2959)
* fix(memory-ingest): --scan-secrets scans the rendered page and fails closed

--scan-secrets ran gitleaks on the raw transcript .jsonl, then imported a
page rendered from it. gitleaks' assignment rules don't match across a
JSON-escaped quote (KEY=\"v\" on disk), so a secret the rendered page
shows as KEY="v" was imported unflagged. And the gate skipped a file only
on scanner "gitleaks" with findings, so a scan that errored (non-zero
exit, 16MB maxBuffer overflow on a file with many findings, unparseable
report) or could not run (gitleaks missing, slow-probe cooldown) imported
the file unscanned.

Scan the rendered page body, the exact bytes writeStaged() writes, via a
new secretScanText() helper, and skip the file whenever the scan did not
complete. Skipped files stay out of the state file, so the next run
retries them. Reword the helper warnings and setup-gbrain/memory.md,
which described the fail-open as intended.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

* fix(test): reconcile Bun failure markers and footer counts

* fix(sync-gbrain): verify source-scoped reads without mutation

* fix(test): recognize grounded TTHW target choices structurally

* fix(aside): make the readiness probe work under zsh and report why it failed

The probe built its deadline into `_T` and expanded it unquoted, so
`$_T aside repl …` only worked in a shell that word-splits. zsh does not: it
looked for a command literally named "gtimeout 30", the probe answered
ASIDE_NOT_RUNNING with Aside installed and ready, and every browsing skill
fell back to the bundled Chromium in silence. zsh is the macOS default and
Aside is macOS-only, so on a stock Mac the probe could never report READY.

The deadline becomes a function, `_gs_d`. It receives the command as "$@",
already split, so sh, bash and zsh all behave the same, and the gtimeout →
timeout → perl alarm chain is unchanged. A 4th arm runs the call unbounded
when none of the three is present, which is what the empty `_T` did before.
Not `eval`: it re-parses the string, so the parens and `;` of the perl arm
become syntax and that arm dies in bash *and* zsh — on a stock Mac, the arm
that actually runs.

On failure the probe now prints the CLI's reason after ASIDE_NOT_RUNNING:,
the shape gstack-render already uses: the first line that starts with a
capital letter, i.e. the CLI's own sentence or Node's `Error:` line below its
loader frame. "Not running" covers states with different fixes — no window
open for the profile, a NODE_OPTIONS preload that kills the CLI — and a bare
verdict sent all of them to "open the Aside app". The BROWSER SETUP prose
quotes that reason before asking the user to open the app.

The text pin asserted the broken invocation verbatim, so it now pins the
function and asserts neither `$_T aside repl` nor an eval form comes back. A
second test executes the rendered probe in sh, bash and zsh on each of the
four deadline arms with stubbed binaries on a narrowed PATH, plus two failing
CLIs: one that prints its own sentence, one that crashes like Node with the
useful line below the frame.

The deadline function costs zero bytes against the lines it replaces; the
reason costs 53 per copy of the probe (44 where the reworded BROWSER SETUP
line gives 9 back). That moves four guards by the measured amount:
plan-devex-review's skeleton cap to 68,550 (measured 68,544), plan-ceo-review's
skeleton cap to 80,150 (measured 80,111) and union ratio to 1.081 (measured
1.0803), and plan-eng-review's union ratio to 1.151 (measured 1.1504).

Fixes #2842, #2941.

* Clarify engineering review startup and decision flow

* Fix Windows readiness fixture PATH and command shim

* fix(test): recognize grounded TTHW target choices structurally

* Clarify engineering review startup and decision flow

* fix(test): restrict QA-only fixture tools to its no-Edit contract

* v1.90.0.0 fix(sync-gbrain): guard readiness verdicts and refresh metadata

* fix(browse): validate canonical upload targets

* fix(gbrain): classify structured PGLite busy response

* fix(browse): preserve native extension runtime APIs

* Fix displayless browser handoff ownership

* Accept unique installed autoplan methodology aliases

* fix(skills): preserve positional literals during installation

* fix(browse): checksum installer contents through stdin

* fix(test): normalize Windows checksum fixture paths

* test: emulate unavailable shasum in Windows checksum fixture

* fix(investigate): preserve owned freeze lifecycle

* fix(review): preserve N+1 retry and Red Team completion

* fix: bound Aside readiness and preserve safe fallback

* test: exercise setup and Chromium on native ARM

* fix: preserve install ownership and ARM browser selection

* Fix gbrain ingest scan boundaries and seed observation

* Refresh managed ship hooks and supervise expanded paid census

* Reject resumed gbrain pages excluded by current policy

* Recover zombie agent locks safely and enable CI Python venv

* Repair paid actor declarations and Aside pitch assertions

* Bump consolidated wave to next free minor release

* Clarify CEO review admin choices and option tradeoffs

* Preserve CEO mode handoff anchors in clarified workflow

* Make Windows portability fixtures use shell-native paths

* Restore ARM Bun alias and clarify ship review gates

* Refresh ship workflow golden snapshots

* Fix Windows DX documentation controls without piped stdin

* Decode Codex child pipes without Bun's encoded-stream stall

* Bound DX pre-review audit before product questions

* Clarify trusted review-start read in paid revalidation

* Bump consolidated wave to next free minor release

* Clarify CEO review admin choices and option tradeoffs

* Preserve CEO mode handoff anchors in clarified workflow

* Make Windows portability fixtures use shell-native paths

* Restore ARM Bun alias and clarify ship review gates

* Refresh ship workflow golden snapshots

* Fix Windows DX documentation controls without piped stdin

* Decode Codex child pipes without Bun's encoded-stream stall

* Bound DX pre-review audit before product questions

* Clarify trusted review-start read in paid revalidation

* Reconcile new main planning flow and paid judge census

* fix: reconcile rebased planning and source-bound validation

* test: pin cookie workflow judge to scored Sonnet model

* fix: keep terminal agent boot out of module imports

* fix: preserve pending-question uncertainty in engineering review

* fix: stabilize Windows reliability-wave fixtures

* fix: clarify design consultation research workflow

* fix: preserve independent design consultation inputs

* fix: resolve design taste scope and browser research guidance

* fix: make consultation opt-in preflight unambiguous

* test: await native Edge owner readiness or terminal result

---------

Co-authored-by: Bruce Krysiak <brucek@alum.mit.edu>
Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
Co-authored-by: Antonio Vitalic <antoninte99@gmail.com>
2026-09-26 18:57:53 -04:00

378 lines
15 KiB
YAML

name: Periodic Evals
# The weekly coverage contract: EVERY periodic-tier paid test runs (EVALS_ALL,
# minus the reasoned excludes in test/helpers/periodic-exclude-data.ts), so
# tests can't rot invisibly — the class where the autoplan-dual-voice E2E was
# silently broken for months until a lucky local diff selected it. Engine:
# scripts/test-paid-shards.ts (the same runner local eval:bg:periodic uses):
# one planner manifest, 6 ordinary slices plus overlay and Autoplan slices, and a FAIL-CLOSED report — a slice
# whose artifact never landed is a failure, not an absence. The gate-census
# job is the weekly EVALS_ALL backstop for the gate tier (PR lanes are
# diff-billed, so without it the full gate census might never execute
# anywhere); the hollow-shard guard (exit 0 + zero executed tests under
# EVALS_ALL fails) makes both lanes census-health checks, not just test runs.
on:
schedule:
- cron: '0 6 * * 1' # Monday 6 AM UTC (ci-image prebuilds at 4 AM)
workflow_dispatch:
concurrency:
group: evals-periodic
cancel-in-progress: true
env:
IMAGE: ghcr.io/${{ github.repository }}/ci
EVALS_PROFILE: full
EVALS_FRESH: "1"
EVALS_CACHE_PURPOSE: periodic
jobs:
build-image:
runs-on: ubicloud-standard-8
timeout-minutes: 15
permissions:
contents: read
packages: write
outputs:
image-tag: ${{ steps.meta.outputs.tag }}
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
- id: meta
# Keep in sync with evals.yml — key on Dockerfile + lockfile only
# (package.json's version field would bust the key on every ship).
# Byte-identity pinned by test/ci-image-tag-binding.test.ts.
run: echo "tag=${{ env.IMAGE }}:${{ hashFiles('.github/docker/Dockerfile.ci', 'bun.lock', 'patches/**') }}" >> "$GITHUB_OUTPUT"
- uses: docker/login-action@dbcb813823bdd20940b903addbd779551569679f # v4
with:
registry: ghcr.io
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
- name: Check if image exists
id: check
run: |
if docker manifest inspect ${{ steps.meta.outputs.tag }} > /dev/null 2>&1; then
echo "exists=true" >> "$GITHUB_OUTPUT"
else
echo "exists=false" >> "$GITHUB_OUTPUT"
fi
- if: steps.check.outputs.exists == 'false'
run: cp package.json bun.lock .github/docker/ && cp -R patches .github/docker/patches
# Registry cache export needs a docker-container builder — the default
# `docker` driver hard-errors on cache-to.
- if: steps.check.outputs.exists == 'false'
uses: docker/setup-buildx-action@37fe631027851001ddb9b187196cc803df7f5f0e # v4
- if: steps.check.outputs.exists == 'false'
uses: docker/build-push-action@53b7df96c91f9c12dcc8a07bcb9ccacbed38856a # v7
with:
context: .github/docker
file: .github/docker/Dockerfile.ci
push: true
# Cron-triggered in the base repo only, so cache export is always safe here.
cache-from: type=registry,ref=${{ env.IMAGE }}:buildcache
cache-to: type=registry,ref=${{ env.IMAGE }}:buildcache,mode=max
tags: |
${{ steps.meta.outputs.tag }}
${{ env.IMAGE }}:latest
plan-slices:
runs-on: ubicloud-standard-8
timeout-minutes: 10
permissions:
contents: read
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
persist-credentials: false
- uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6 # v2
with:
bun-version: 1.4.0
- name: Emit run manifest (ALL periodic tests minus reasoned excludes)
env:
EVALS_ALL: "1"
run: EVALS_TIER=periodic bun --no-install run scripts/test-paid-shards.ts --tier periodic --emit-plan /tmp/paid-plan/manifest.json --slices 8 --autoplan-slice
- uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
with:
name: paid-plan
path: /tmp/paid-plan/manifest.json
retention-days: 30
- name: Emit gate census manifest (ALL gate tests)
env:
EVALS_ALL: "1"
run: EVALS_TIER=gate bun run scripts/test-paid-shards.ts --tier gate --emit-plan /tmp/gate-census-plan/manifest.json --slices 7
- uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
with:
name: gate-census-plan
path: /tmp/gate-census-plan/manifest.json
retention-days: 30
eval-slices:
runs-on: ubicloud-standard-8
needs: [build-image, plan-slices]
# Eight slices retain every registered case and retry. The complete
# census needs at most 338 minutes per slice, plus 20 minutes setup/upload.
timeout-minutes: 358
permissions:
contents: read
packages: read
container:
image: ${{ needs.build-image.outputs.image-tag }}
credentials:
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
options: --user runner
strategy:
fail-fast: false
matrix:
slice: [1, 2, 3, 4, 5, 6, 7, 8]
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
# Full history: files with SELF-derived selection (the LLM-judge
# map, routing) walk git at module load, and selection is
# fail-closed on git errors — a shallow checkout crashed those
# shards on the lane's first live run ("ambiguous argument
# 'main...HEAD'"). The manifest still governs WHICH shards run.
fetch-depth: 0
persist-credentials: false
- name: Fix bun temp
uses: ./.github/actions/fix-bun-temp
- name: Restore deps
uses: ./.github/actions/restore-deps
- run: bun run build
# Any slice can host a PTY test — seed + registration run
# unconditionally (idempotent; mirrors evals.yml's sliced lane). The
# register composite carries the fail-fast dangling-symlink/frontmatter
# verification loop — this lane previously LACKED it, so a moved skill
# target surfaced as a silent "Unknown command" + wedged PTY session.
- name: Seed claude interactive config
uses: ./.github/actions/seed-claude-config
with:
anthropic-api-key: ${{ secrets.ANTHROPIC_API_KEY }}
- name: Register gstack skills for PTY tests
uses: ./.github/actions/register-gstack-skills
- uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8
with:
name: paid-plan
path: /tmp/paid-plan
- name: Run slice ${{ matrix.slice }}/8
env:
ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
PLAYWRIGHT_BROWSERS_PATH: /opt/playwright-browsers
EVALS_JOBS: "2"
EVALS_CONCURRENCY: "2"
GSTACK_EVAL_DIR: /tmp/paid-slice-results
run: EVALS_TIER=periodic bun run scripts/test-paid-shards.ts --tier periodic --plan /tmp/paid-plan/manifest.json --slice ${{ matrix.slice }}
- name: Upload slice results
if: always()
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
with:
name: paid-slice-${{ matrix.slice }}
path: /tmp/paid-slice-results
retention-days: 90
- name: Upload shard logs on failure
if: failure()
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
with:
name: paid-slice-${{ matrix.slice }}-logs
include-hidden-files: true
# The Fix-bun-temp step points TMPDIR at /home/runner/.cache, so the
# runner's spool lands THERE, not /tmp — the original /tmp glob
# uploaded nothing and a red slice's diagnostics were unreachable.
path: |
/home/runner/.cache/gstack-paid-shard-*.log
/tmp/gstack-paid-shard-*.log
if-no-files-found: ignore
retention-days: 30
# Weekly EVALS_ALL gate-tier census: PR lanes are diff-billed, so without
# this the full gate census might never execute anywhere and the selector's
# blind spots rot invisibly. Census health, not selector correctness —
# selector logic has free synthetic-diff contract tests.
gate-census:
runs-on: ubicloud-standard-8
needs: [build-image, plan-slices]
# Seven slices need at most 304m each, plus 20 minutes setup/upload.
timeout-minutes: 352
permissions:
contents: read
packages: read
container:
image: ${{ needs.build-image.outputs.image-tag }}
credentials:
username: ${{ github.actor }}
password: ${{ secrets.GITHUB_TOKEN }}
options: --user runner
strategy:
# Four file workers total, each retaining two in-file case workers.
fail-fast: false
max-parallel: 4
matrix:
slice: [1, 2, 3, 4, 5, 6, 7]
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
# Full history: files with SELF-derived selection (the LLM-judge
# map, routing) walk git at module load, and selection is
# fail-closed on git errors — a shallow checkout crashed those
# shards on the lane's first live run ("ambiguous argument
# 'main...HEAD'"). The manifest still governs WHICH shards run.
fetch-depth: 0
persist-credentials: false
- name: Fix bun temp
uses: ./.github/actions/fix-bun-temp
- name: Restore deps
uses: ./.github/actions/restore-deps
- run: bun run build
- uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8
with:
name: gate-census-plan
path: /tmp/gate-census-plan
- name: Run gate census slice ${{ matrix.slice }}/7
env:
ANTHROPIC_API_KEY: ${{ secrets.ANTHROPIC_API_KEY }}
OPENAI_API_KEY: ${{ secrets.OPENAI_API_KEY }}
GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
PLAYWRIGHT_BROWSERS_PATH: /opt/playwright-browsers
EVALS_JOBS: "1"
EVALS_CONCURRENCY: "2"
GSTACK_EVAL_DIR: /tmp/gate-census-results
run: EVALS_TIER=gate bun run scripts/test-paid-shards.ts --tier gate --plan /tmp/gate-census-plan/manifest.json --slice ${{ matrix.slice }}
- name: Upload census results
if: always()
uses: actions/upload-artifact@043fb46d1a93c77aae656e7c1c64a875d1fc6a0a # v7
with:
name: gate-census-${{ matrix.slice }}
path: /tmp/gate-census-results
retention-days: 90
report:
runs-on: ubicloud-standard-2
needs: [plan-slices, eval-slices, gate-census]
# always(): the report must run (and FAIL) when an executor died — a
# missing slice artifact reading as green is the class this lane kills.
if: always() && needs.plan-slices.result == 'success'
timeout-minutes: 10
permissions:
contents: read
# The failure notification below upserts a tracking issue via
# `gh api /issues` — gated by the issues permission.
issues: write
steps:
- uses: actions/checkout@3d3c42e5aac5ba805825da76410c181273ba90b1 # v7
with:
persist-credentials: false
- uses: oven-sh/setup-bun@0c5077e51419868618aeaa5fe8019c62421857d6 # v2
with:
bun-version: 1.4.0
- uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8
with:
name: paid-plan
path: /tmp/paid-report
- uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8
with:
pattern: paid-slice-[0-9]*
path: /tmp/paid-report
merge-multiple: true
- uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8
with:
name: gate-census-plan
path: /tmp/gate-census-report
- uses: actions/download-artifact@3e5f45b2cfb9172054b4087a40e8e0b5a5461e7c # v8
with:
pattern: gate-census-[0-9]*
path: /tmp/gate-census-report
merge-multiple: true
- name: Reconcile slices against the manifest (fail-closed)
id: reconcile
if: always()
run: |
set +e
EVALS_TIER=periodic bun --no-install run scripts/test-paid-shards.ts --tier periodic --report /tmp/paid-report | tee /tmp/report.txt
# PIPESTATUS[0], NOT $?: GitHub's default run-step shell is
# `bash -e {0}` with NO pipefail, so $? after the pipe is tee's
# exit (always 0) — the fail-closed gate was silently fail-open
# (caught by the ship review army; the wiring test now pins this).
echo "exit=${PIPESTATUS[0]}" >> "$GITHUB_OUTPUT"
- name: Reconcile gate census against the manifest (fail-closed)
id: gate-reconcile
if: always()
run: |
set +e
EVALS_TIER=gate bun run scripts/test-paid-shards.ts --tier gate --report /tmp/gate-census-report | tee /tmp/gate-report.txt
echo "exit=${PIPESTATUS[0]}" >> "$GITHUB_OUTPUT"
# A red weekly lane nobody must action is waste — upsert ONE tracking
# issue (never a new issue per week) with the reconciliation output, so
# failures have an owner-visible artifact with history in one place.
- name: Upsert tracking issue on failure
if: always() && (steps.reconcile.outputs.exit != '0' || steps.gate-reconcile.outputs.exit != '0' || needs.eval-slices.result != 'success' || needs.gate-census.result != 'success')
env:
GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
run: |
set -euo pipefail
TITLE="Weekly periodic evals: red lane needs triage"
BODY_FILE=/tmp/issue-body.md
{
echo "Automated weekly report — run: ${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID}"
echo
echo "- periodic reconciliation exit: ${{ steps.reconcile.outputs.exit }}"
echo "- periodic slices job: ${{ needs.eval-slices.result }}"
echo "- gate census reconciliation exit: ${{ steps.gate-reconcile.outputs.exit }}"
echo "- gate census job: ${{ needs.gate-census.result }}"
echo
echo '```'
tail -c 6000 /tmp/report.txt 2>/dev/null || echo "(no reconciliation output)"
echo '```'
echo
echo '```'
tail -c 6000 /tmp/gate-report.txt 2>/dev/null || echo "(no gate census reconciliation output)"
echo '```'
echo
echo "Exclusion policy: test/helpers/periodic-exclude-data.ts (every entry needs reason + tracking; removal re-activates the file next week)."
} > "$BODY_FILE"
EXISTING=$(gh issue list --repo "$GITHUB_REPOSITORY" --state open --search "in:title \"$TITLE\"" --json number --jq '.[0].number // empty')
if [ -n "$EXISTING" ]; then
gh issue comment "$EXISTING" --repo "$GITHUB_REPOSITORY" --body-file "$BODY_FILE"
echo "commented on #$EXISTING"
else
gh issue create --repo "$GITHUB_REPOSITORY" --title "$TITLE" --body-file "$BODY_FILE"
fi
- name: Fail the workflow when reconciliation failed
if: always() && (steps.reconcile.outputs.exit != '0' || steps.gate-reconcile.outputs.exit != '0' || needs.eval-slices.result != 'success' || needs.gate-census.result != 'success')
run: exit 1