callstack__agent-device/.github/workflows/mutation-weekly.yml

name: Mutation Weekly

# Weekly mutation sweep over the enumerated decision kernels (issue #1415).
# Mutation score is the mechanical answer to "is this test load-bearing or
# decorative"; a full-suite sweep is unaffordable, so the scope is the kernel
# registry in scripts/mutation/modules.ts and nothing else.
#
# Sharded one job per module: the whole sweep is ~2,150 mutants, and the selector
# module alone is ~1,280 of them, so a single job would sit near the 30-minute
# acceptance budget on an ubuntu runner. The shards' JSON reports are merged into
# one verdict by the ratchet job (`--report-dir`), so the ratchet still sees a
# full sweep.
#
# The lane reports; it does not commit. The proposed baseline rides in the
# artifact and applying it is a reviewed `pnpm mutation:baseline` commit — a score
# can never lower itself. Gating (after two consecutive stable weekly runs) is
# enforced on PRs by mutation-affected.yml.

on:
  schedule:
    # Sundays 05:00 UTC — after the nightly lanes, before the working week.
    - cron: '0 5 * * 0'
  workflow_dispatch:

permissions:
  contents: read

concurrency:
  group: ci-${{ github.workflow }}-${{ github.ref }}
  cancel-in-progress: true

jobs:
  shard:
    name: Mutants (${{ matrix.name }})
    runs-on: ubuntu-latest
    # The acceptance budget is 30 minutes of wall clock for the lane, and shards
    # run in parallel, so the budget is per shard. The observed rate on a 2-core
    # runner is ~3s/mutant, which is why selectors (~1,280 mutants) is sliced.
    timeout-minutes: 30
    strategy:
      fail-fast: false
      matrix:
        # Kept in step with KERNEL_MODULES by the shard-coverage assertion in
        # scripts/mutation/workflow.test.ts, which also pins --expect-shards to
        # the number of entries here.
        include:
          - { name: kernel-errors, module: kernel-errors }
          - { name: daemon-ref-frame, module: daemon-ref-frame }
          - { name: interaction-settle, module: interaction-settle }
          - { name: scroll-edge-state, module: scroll-edge-state }
          - { name: selectors-1, module: selectors, shard: 1/4 }
          - { name: selectors-2, module: selectors, shard: 2/4 }
          - { name: selectors-3, module: selectors, shard: 3/4 }
          - { name: selectors-4, module: selectors, shard: 4/4 }
    steps:
      - name: Checkout
        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      - name: Setup toolchain
        uses: ./.github/actions/setup-node-pnpm

      - name: Run mutants for ${{ matrix.name }}
        run: |
          pnpm mutation:run --modules ${{ matrix.module }} \
            ${{ matrix.shard && format('--shard {0}', matrix.shard) || '' }}

      # A shard can die before Stryker writes anything (install, config, crash);
      # without this the artifact is absent and "shard failed" is indistinguishable
      # from "lane never ran". --fail-envelope leaves a real verdict untouched.
      - name: Record a failed shard envelope
        if: failure()
        run: |
          pnpm mutation:run --modules ${{ matrix.module }} --fail-envelope \
            "shard ${{ matrix.name }} failed before producing a report (run ${{ github.run_id }})" || true

      - name: Upload shard report
        if: always()
        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
        with:
          name: mutation-shard-${{ matrix.name }}
          path: |
            .tmp/mutation/mutation.json
            .tmp/mutation/lane-envelope.json
          if-no-files-found: warn

  ratchet:
    name: Mutation ratchet
    runs-on: ubuntu-latest
    needs: shard
    if: always()
    timeout-minutes: 15
    steps:
      - name: Checkout
        uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2

      - name: Setup toolchain
        uses: ./.github/actions/setup-node-pnpm

      - name: Ratchet self-test
        run: pnpm mutation:test

      - name: Download shard reports
        uses: actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093 # v4.3.0
        with:
          pattern: mutation-shard-*
          path: .tmp/mutation/shards

      # run.ts writes the markdown verdict to $GITHUB_STEP_SUMMARY when the
      # runner exports it, so the summary and the artifact carry the same numbers.
      - name: Ratchet the merged sweep and propose the next baseline
        run: |
          pnpm mutation:check --report-dir .tmp/mutation/shards --expect-shards 8 --update
          cp mutation-baselines/decision-kernels.json .tmp/mutation/proposed-baseline.json
          git checkout -- mutation-baselines/decision-kernels.json

      # The self-test and the artifact download both run before the ratchet, so a
      # failure there would otherwise leave the aggregate lane with no envelope.
      - name: Record a failed lane envelope
        if: failure()
        run: |
          pnpm mutation:run --fail-envelope \
            "weekly ratchet job failed before producing a verdict (run ${{ github.run_id }})" || true

      # Freshness/drift telemetry (#1430): the envelope states commit, Stryker
      # version, config hash, duration and result, so a lane going dark or a tool
      # bump is visible without reading these logs.
      - name: Lane envelope
        if: always()
        run: |
          if [ ! -f .tmp/mutation/lane-envelope.json ]; then
            echo 'No lane envelope was written — the job died before it could run node.' \
              >> "$GITHUB_STEP_SUMMARY"
            exit 0
          fi
          {
            echo '<details><summary>Lane envelope (schema #1430)</summary>'
            echo
            echo '```json'
            cat .tmp/mutation/lane-envelope.json
            echo '```'
            echo '</details>'
          } >> "$GITHUB_STEP_SUMMARY"

      - name: Upload mutation report
        if: always()
        uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4.6.2
        with:
          name: mutation-decision-kernels
          path: |
            .tmp/mutation/shards
            .tmp/mutation/lane-envelope.json
            .tmp/mutation/proposed-baseline.json
          if-no-files-found: warn