Preserve label_command activation guards when frontmatter defines a custom if:

[Copilot]

label_command workflows were dropping their compiler-derived label match guard whenever the workflow also defined a top-level frontmatter if:. That caused activation to widen from “matching label only” to “any label event that satisfies the custom condition”.

• Compiler behavior

  • Change label_command compilation to always AND-combine the derived label-trigger condition with any existing workflow if:.
  • Preserve the intended activation shape:
    • pre-activation gate
    • trigger-derived label match
    • user-authored custom condition

• Regression coverage

  • Add a focused test for a label_command workflow with a custom top-level if:.
  • Assert the compiled activation job still contains both the label-name check and the custom predicate.

• Result

  • Custom workflow conditions now refine label_command activation instead of replacing it.

Example of the compiled behavior after this change:

if: >-
  ${{ needs.pre_activation.outputs.activated == 'true'
      && (github.event_name == 'issues' && github.event.label.name == 'state:approved')
      && (github.event.issue.state == 'open') }}

• Fixes Custom frontmatter if: replaces the label_command-derived condition instead of combining with it #47594

---

Generated by 👨‍🍳 PR Sous Chef · gpt54 · 10.9 AIC · ⌖ 8.52 AIC · ⊞ 7.1K · ◷
_{Comment /souschef to run again}

---

Generated by 👨‍🍳 PR Sous Chef · gpt54 · 14 AIC · ⌖ 8.18 AIC · ⊞ 7.1K · ◷
_{Comment /souschef to run again}

[Copilot]

Pull request overview

Fixes label_command compilation so custom conditions refine rather than replace label guards.

Changes:

• AND-combines label and custom conditions.
• Adds regression coverage.
• Registers the release-workflow guidance file.

Show a summary per file

File	Description
pkg/workflow/tools.go	Combines label guards with existing conditions.
pkg/workflow/label_command_test.go	Tests compiled activation conditions.
.github/skills/agentic-workflows/SKILL.md	Adds release-workflow guidance.

Review details

Tip

Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

• Files reviewed: 3/3 changed files
• Comments generated: 1
• Review effort level: Medium

[github-actions]

🤖 PR Triage — §30192186368

Field	Value
Category	bug
Risk	🟡 medium
Score	68 / 100 (impact 38 + urgency 18 + quality 12)
Action	⚡ fast_track

Why: Fixes #47594 — label_command custom if: silently replaced the compiler-derived label match guard instead of AND-combining, widening activation beyond intent. Copilot reviewer check passing, focused regression test added.

Next: Full CI confirmation, then fast-track merge.

Generated by 🔧 PR Triage Agent · sonnet46 · 48.8 AIC · ⌖ 8.75 AIC · ⊞ 5.7K · ◷

[gh-aw-bot]

@copilot please run the pr-finisher skill, address the unresolved review feedback below, and rerun checks once the branch is up to date.

Unresolved review feedback:

• Copilot: Preserve label_command activation guards when frontmatter defines a custom if: #48105 (comment)

Generated by 👨‍🍳 PR Sous Chef · gpt54 · 10.9 AIC · ⌖ 8.52 AIC · ⊞ 7.1K · ◷
_{Comment /souschef to run again}

[Copilot]

@copilot please run the pr-finisher skill, address the unresolved review feedback below, and rerun checks once the branch is up to date.

Unresolved review feedback:...

Addressed the unresolved thread in commit c9974aea1a (plus bde56a1236).

Changes made:

• pkg/workflow/tools.go: always AND-combine user if: with the derived command/label trigger guard in applyCommandTriggerOnSection (including mixed slash_command + label_command workflows).
• pkg/workflow/label_command_test.go: added mixed-trigger regression coverage with a custom top-level if: and assertion for both slash and label guards.

Checks rerun locally: targeted workflow tests pass; make fmt, make lint, make test-unit, and make recompile were run. make test still has a pre-existing unrelated failure in TestFallbackAWFilesMatchesLocalAWDirectory (fallback list drift). Since cloud-agent pushes do not trigger CI, a maintainer re-trigger is still required for fresh PR checks on the new HEAD.

[gh-aw-bot]

@copilot please run the pr-finisher skill, confirm the unresolved review feedback is fully addressed on the latest head, and rerun checks once the branch is up to date.

Unresolved review feedback:

• Copilot: Preserve label_command activation guards when frontmatter defines a custom if: #48105 (comment)

Generated by 👨‍🍳 PR Sous Chef · gpt54 · 14 AIC · ⌖ 8.18 AIC · ⊞ 7.1K · ◷
_{Comment /souschef to run again}

[Copilot]

@copilot please run the pr-finisher skill, confirm the unresolved review feedback is fully addressed on the latest head, and rerun checks once the branch is up to date.

Unresolved review feedback:...

Confirmed on latest HEAD (634474d) that the unresolved Copilot review feedback is fully addressed:

• applyCommandTriggerOnSection now AND-combines custom if: with the derived mixed trigger guard (slash_command OR label_command) in /home/runner/work/gh-aw/gh-aw/pkg/workflow/tools.go.
• Mixed-trigger regression coverage with custom top-level if: is present in /home/runner/work/gh-aw/gh-aw/pkg/workflow/label_command_test.go (TestSlashAndLabelCommandWorkflowCompileCombinesCustomIf).

Checks rerun locally on this branch:

• make fmt ✅
• make lint ✅
• make test-unit ✅
• make test ✅
• make recompile ✅

I also reviewed prior PR CI logs via GitHub MCP. The historical failure was the known fallback AW file drift test; current PR-head CI runs are stale/action_required after agent push and require a maintainer-triggered rerun.

[github-actions]

🎉 This pull request is included in a new release.

Release: v0.83.4