Repository navigation
cherry-pick(release/v1.206): run a package with no project in the default project (#4630) #6751
Workflow file for this run
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Validate skills.sh Grouping | |
| on: | |
| # No `paths:` filter on purpose, for the same two reasons as | |
| # validate-version-sync.yml and smoke-rpa-skills.yml: | |
| # | |
| # 1. Required-check safety: if `paths:` excluded a PR the workflow wouldn't | |
| # run, no check would be reported, and merge would block forever. The job | |
| # is a checkout plus a ~20s dependency-free script, so just always run it. | |
| # 2. A path list is not a reliable trigger for this drift. The filter that | |
| # used to live here listed `skills/*/SKILL.md`, yet the check never ran on | |
| # #2252 — the PR that introduced the very drift this guard exists to | |
| # catch — so the guard was silent exactly when it mattered. | |
| pull_request: | |
| workflow_dispatch: | |
| permissions: | |
| contents: read | |
| jobs: | |
| validate-skills-sh: | |
| runs-on: uipath-ubuntu-latest | |
| name: Validate skills.sh.json against skills/ | |
| steps: | |
| - uses: actions/checkout@11d5960a326750d5838078e36cf38b85af677262 # v4.4.0 | |
| with: | |
| # On a pull_request event this checks out refs/pull/N/merge, whose | |
| # first parent is the base commit. Depth 2 is all the baseline diff | |
| # needs — no unshallowing of the full history. | |
| fetch-depth: 2 | |
| - uses: actions/setup-python@a26af69be951a213d495a4c3e4e4022e16d87065 # v5.6.0 | |
| with: | |
| python-version: '3.13' | |
| # skills.sh.json only affects the repo's page on skills.sh, so nothing | |
| # else fails when it drifts. Adding or renaming a skill without updating | |
| # it leaves the public page showing a stale grouping — this is the only | |
| # thing that notices. | |
| # | |
| # --baseline-ref keeps the whole-tree read (drift stays visible) but | |
| # scopes the exit code to drift THIS PR introduces. Without it, one | |
| # ungrouped skill on main fails every unrelated PR that touches any | |
| # skills/*/SKILL.md: #2252 landed uipath-process-mining ungrouped on | |
| # 2026-08-04 and cost ~20 unrelated branches 65 red runs over 48 hours. | |
| # | |
| # HEAD^1 is the base parent of the merge commit actually being tested, so | |
| # the baseline always matches the tree under test. The event payload's | |
| # `pull_request.base.sha` is the base tip as of the triggering event and | |
| # can lag the merge ref GitHub recomputed. | |
| # | |
| # A manual dispatch has no merge commit, so it gets no baseline and | |
| # validates the whole tree strictly — which is what you want when asking | |
| # "is the tree clean right now?". | |
| - name: Validate skills.sh grouping | |
| env: | |
| IS_PR: ${{ github.event_name == 'pull_request' }} | |
| run: | | |
| if [ "$IS_PR" = "true" ]; then | |
| python3 scripts/check-skills-sh.py --baseline-ref HEAD^1 | |
| else | |
| python3 scripts/check-skills-sh.py | |
| fi |