You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
[H-CI] Verify contributor fork PR flow and roll out repository-specific checks #37
The archived CI/CD PR Readiness status note identified an organization-wide contributor flow and type-specific verification gap. Its claim that navigation PR #5 was awaiting first-contributor approval is obsolete: PR #5 is merged, and the interface-only H-CI issue #6 is closed with combined evidence. This issue tracks the remaining broader work, without reopening completed interface checks.
Scope
Prove an external fork PR can open, run the relevant build/tests, display clear pass/fail evidence, and reach the correct reviewer.
Add repository-specific build/test gates and a matching local verification command to the priority repositories: openamr-platform-sw, openamrobot-manipulation, and any remaining interface integration outside the closed issue Update README.md #6.
Keep generated artifacts and test evidence discoverable on PRs; avoid skipped or weakened checks producing green status.
Record one successful fork-PR walkthrough with workflow approval behavior and reviewer routing, then link the repository-specific gate PRs and their failing/passing evidence. Owners and due dates should be confirmed by the team; the archived note listed Arumuga for CI conventions and Karthikeyan for release/install.
image PR #41 remains awaiting code-owner review; canonical GHCR digest update is a post-merge dependency.
Acceptance gap deliberately open: these are same-repository or existing contributor PR runs, not proof of an external fork PR. An outside collaborator must fork one priority repository, open a harmless sample PR, confirm fork workflow approval and read-only credentials, inspect the actual non-skipped build/tests/artifacts, and verify CODEOWNERS reviewer routing. No external fork was created by this audit, and no such end-to-end evidence is claimed. Release install/consumer validation and hardware acceptance remain separate from this issue.
31 October CI integration update: openamrobot-interfaces PR #29 is merged to main, squash SHA 14912e18ecb6473ebd7ef4a42c4bd1870b2023f6. The shared quality/test gate is now enabled for contributor PRs rather than skipped. Before merge it passed shared hosted run 40887451839 and dedicated ROS Jazzy run 40887451149: 15 schema tests, 14 compatibility tests, 13 generated types, consumer + both middleware exchange checks. Mohammad and Karthikeyan approved, DCO green. Merge method was repository-supported squash.
Still open: platform-sw #101 (newest commit DCO identity mismatch and dismissed prior reviews; see PR comment), then platform-sw #57 after #101 lands; image #41 awaits Alex's merge approval, followed by main GHCR publish/new digest. No actual external-fork end-to-end pilot or contributor-machine GUI VS Code startup yet. Alex's separate security harness PR #70 has a focused, tested follow-up PR #76 for review; no activation authorized.
Activity
wikki26 commentedon Oct 29, 2026
Contributor-gate rollout evidence (29 Oct 2026, GitHub-hosted):
quality/testthrough reusable CI. Run 40884563081: PASS, 8 ROS packages built, 169 tests, 0 failures, 6 tracked skips, artifacts retained. Pending maintainer approval/merge.postCreateCommandanddevcontainer execvalidated. Its tested branch includes #101; wait for #101 and canonical image ci: add baseline ROS 2 CI image #41 before final merge, and still obtain actual graphical VS Code startup evidence.maincaller still lacksverify: true, so verify that rollout separately when Enhance README with demo and maintainer details #20 lands.Acceptance gap deliberately open: these are same-repository or existing contributor PR runs, not proof of an external fork PR. An outside collaborator must fork one priority repository, open a harmless sample PR, confirm fork workflow approval and read-only credentials, inspect the actual non-skipped build/tests/artifacts, and verify CODEOWNERS reviewer routing. No external fork was created by this audit, and no such end-to-end evidence is claimed. Release install/consumer validation and hardware acceptance remain separate from this issue.
wikki26 commentedon Oct 31, 2026
31 October CI integration update: openamrobot-interfaces PR #29 is merged to main, squash SHA
14912e18ecb6473ebd7ef4a42c4bd1870b2023f6. The sharedquality/testgate is now enabled for contributor PRs rather than skipped. Before merge it passed shared hosted run 40887451839 and dedicated ROS Jazzy run 40887451149: 15 schema tests, 14 compatibility tests, 13 generated types, consumer + both middleware exchange checks. Mohammad and Karthikeyan approved, DCO green. Merge method was repository-supported squash.Still open: platform-sw #101 (newest commit DCO identity mismatch and dismissed prior reviews; see PR comment), then platform-sw #57 after #101 lands; image #41 awaits Alex's merge approval, followed by main GHCR publish/new digest. No actual external-fork end-to-end pilot or contributor-machine GUI VS Code startup yet. Alex's separate security harness PR #70 has a focused, tested follow-up PR #76 for review; no activation authorized.