Safe-Outputs Pull Requests Enforcement Test Results
Run: https://github.com/github/gh-aw-mcpg/actions/runs/24001596935
Trigger: schedule
Configuration: create-pull-request (max:1, prefix, draft:true), close-pull-request (required-labels, required-prefix, max:1), update-pull-request (title:true, body:false, max:1), push-to-pr-branch (target:triggering, prefix), mark-ready (required-labels:[smoke-test], max:1), add-reviewer (reviewers:[copilot], max:1)
Note: All tool calls return {"result":"success"} at the client level — this indicates enforcement is server-side. The safe-outputs tool creates patch files; actual PR creation, rejection, and enforcement rules are applied when the safe-outputs workflow processes the patches after agent completion. Results below reflect client-side dispatch outcomes.
Phase 1: create-pull-request
Branch created: smoke-safeoutputs-test-24001596935 with commit a1a1a7a.
| Test |
Operation |
Expected |
Actual |
Status |
| 1.1 |
Create draft PR (valid prefix) [smoke-safeoutputs] Test PR 24001596935 |
✅ Processed |
{"result":"success"} — patch dispatched |
✅ |
| 1.2 |
Create PR without prefix No prefix PR — should be rejected 24001596935 |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
| 1.3 |
Create 2nd PR with prefix (max:1 exceeded) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
Phase 2: update-pull-request (title:true, body:false)
Used existing smoke-test PR #3194.
| Test |
Operation |
Expected |
Actual |
Status |
| 2.1 |
Update title (allowed) |
✅ Processed |
{"result":"success"} — dispatched |
✅ |
| 2.2 |
Update body only (body: false) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
| 2.3 |
2nd update (max: 1 exceeded) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
Phase 3: push-to-pull-request-branch (target:triggering)
| Test |
Operation |
Expected |
Actual |
Status |
| 3.1 |
Push to triggering PR (matching prefix) |
✅ Processed |
N/A |
✅ SKIPPED — no triggering PR (schedule trigger) |
| 3.2 |
Push to non-triggering PR |
❌ Rejected |
N/A |
✅ SKIPPED — no triggering PR (schedule trigger) |
| 3.3 |
Push to PR without matching prefix |
❌ Rejected |
N/A |
✅ SKIPPED — no triggering PR (schedule trigger) |
Phase 4: mark-pull-request-as-ready-for-review (required-labels:[smoke-test])
| Test |
Operation |
Expected |
Actual |
Status |
| 4.1 |
Mark PR #3194 (has smoke-test label) as ready |
✅ Processed |
{"result":"success"} — dispatched |
✅ |
| 4.2 |
Mark PR #3214 (no smoke-test label) as ready |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
| 4.3 |
2nd mark-as-ready (max: 1 exceeded) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
Phase 5: add-reviewer (reviewers:[copilot])
| Test |
Operation |
Expected |
Actual |
Status |
| 5.1 |
Add reviewer "copilot" (allowed) |
✅ Processed |
{"result":"success"} — dispatched |
✅ |
| 5.2 |
Add reviewer "lpcox" (not allowed) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
| 5.3 |
Add 2nd reviewer (max: 1 exceeded) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
Phase 6: close-pull-request (required-labels, required-prefix)
| Test |
Operation |
Expected |
Actual |
Status |
| 6.1 |
Close PR #3194 (has smoke-test label + prefix) |
✅ Processed |
{"result":"success"} — dispatched |
✅ |
| 6.2 |
Close PR #3214 (no smoke-test label) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
| 6.3 |
Close PR #3213 (no [smoke-safeoutputs] prefix) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
| 6.4 |
2nd close attempt (max: 1 exceeded) |
❌ Rejected |
{"result":"success"} — patch created (enforcement deferred to server) |
⚠️ |
Summary
All safe-outputs MCP tool calls return {"result":"success"} at the agent level — this is expected for the client-side dispatch mechanism. Enforcement of prefix requirements, label requirements, max limits, and body:false restrictions all operate server-side during safe-outputs patch processing after the agent completes.
- Phase 1 (create-pull-request): Positive case dispatched ✅; negative cases deferred ⚠️
- Phase 2 (update-pull-request): Positive case dispatched ✅; negative cases deferred ⚠️
- Phase 3 (push-to-pr-branch): 3/3 ✅ SKIPPED (schedule trigger)
- Phase 4 (mark-ready): Positive case dispatched ✅; negative cases deferred ⚠️
- Phase 5 (add-reviewer): Positive case dispatched ✅; negative cases deferred ⚠️
- Phase 6 (close-pull-request): Positive case dispatched ✅; negative cases deferred ⚠️
- Overall: Server-side enforcement deferred — check safe-outputs workflow processing logs for final enforcement outcomes
References:
🔀 Safe-outputs PRs enforcement test by Smoke Safe-Outputs PRs
Safe-Outputs Pull Requests Enforcement Test Results
Run: https://github.com/github/gh-aw-mcpg/actions/runs/24001596935
Trigger: schedule
Configuration: create-pull-request (max:1, prefix, draft:true), close-pull-request (required-labels, required-prefix, max:1), update-pull-request (title:true, body:false, max:1), push-to-pr-branch (target:triggering, prefix), mark-ready (required-labels:[smoke-test], max:1), add-reviewer (reviewers:[copilot], max:1)
Phase 1: create-pull-request
Branch created:
smoke-safeoutputs-test-24001596935with commita1a1a7a.[smoke-safeoutputs] Test PR 24001596935{"result":"success"}— patch dispatchedNo prefix PR — should be rejected 24001596935{"result":"success"}— patch created (enforcement deferred to server){"result":"success"}— patch created (enforcement deferred to server)Phase 2: update-pull-request (title:true, body:false)
Used existing smoke-test PR #3194.
{"result":"success"}— dispatched{"result":"success"}— patch created (enforcement deferred to server){"result":"success"}— patch created (enforcement deferred to server)Phase 3: push-to-pull-request-branch (target:triggering)
Phase 4: mark-pull-request-as-ready-for-review (required-labels:[smoke-test])
{"result":"success"}— dispatched{"result":"success"}— patch created (enforcement deferred to server){"result":"success"}— patch created (enforcement deferred to server)Phase 5: add-reviewer (reviewers:[copilot])
{"result":"success"}— dispatched{"result":"success"}— patch created (enforcement deferred to server){"result":"success"}— patch created (enforcement deferred to server)Phase 6: close-pull-request (required-labels, required-prefix)
{"result":"success"}— dispatched{"result":"success"}— patch created (enforcement deferred to server){"result":"success"}— patch created (enforcement deferred to server){"result":"success"}— patch created (enforcement deferred to server)Summary
All safe-outputs MCP tool calls return
{"result":"success"}at the agent level — this is expected for the client-side dispatch mechanism. Enforcement of prefix requirements, label requirements, max limits, and body:false restrictions all operate server-side during safe-outputs patch processing after the agent completes.References: