Safe-Outputs Pull Requests Enforcement Test Results
Run: https://github.com/github/gh-aw-mcpg/actions/runs/26684163438
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
| Test |
Operation |
Expected |
Actual |
Status |
| 1.1 |
Create draft PR (valid prefix) |
✅ Processed |
Tool returned {"result":"success","patch":{...},"bundle":{...}} — declaration accepted |
✅ |
| 1.2 |
Create PR without prefix |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
| 1.3 |
Create 2nd PR (max exceeded) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
Phase 2: update-pull-request (title:true, body:false)
| Test |
Operation |
Expected |
Actual |
Status |
| 2.1 |
Update title (allowed) |
✅ Processed |
Tool returned {"result":"success"} — declaration accepted |
✅ |
| 2.2 |
Update body (body: false) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
| 2.3 |
2nd update (max: 1 exceeded) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
Phase 3: push-to-pull-request-branch (target:triggering)
| Test |
Operation |
Expected |
Actual |
Status |
| 3.1 |
Push to triggering PR (matching prefix) |
✅ Processed |
SKIPPED - no triggering item (schedule trigger) |
SKIPPED |
| 3.2 |
Push to non-triggering PR |
❌ Rejected |
SKIPPED - no triggering item (schedule trigger) |
SKIPPED |
| 3.3 |
Push to PR without matching prefix |
❌ Rejected |
SKIPPED - no triggering item (schedule trigger) |
SKIPPED |
Phase 4: mark-pull-request-as-ready-for-review (required-labels:[smoke-test])
| Test |
Operation |
Expected |
Actual |
Status |
| 4.1 |
Mark PR with smoke-test label as ready (PR #6718) |
✅ Processed |
Tool returned {"result":"success"} — declaration accepted |
✅ |
| 4.2 |
Mark PR without required label as ready (PR #6716) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
| 4.3 |
2nd mark-as-ready (max: 1 exceeded) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
Phase 5: add-reviewer (reviewers:[copilot])
| Test |
Operation |
Expected |
Actual |
Status |
| 5.1 |
Add reviewer "copilot" (allowed) (PR #6718) |
✅ Processed |
Tool returned {"result":"success"} — declaration accepted |
✅ |
| 5.2 |
Add non-allowed reviewer "octocat" |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
| 5.3 |
Add 2nd reviewer (max: 1 exceeded) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
Phase 6: close-pull-request (required-labels, required-prefix)
| Test |
Operation |
Expected |
Actual |
Status |
| 6.1 |
Close PR with required label+prefix (PR #6718) |
✅ Processed |
Tool returned {"result":"success"} — declaration accepted |
✅ |
| 6.2 |
Close PR without required label (PR #6716) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
| 6.3 |
Close PR without required prefix (PR #6715) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
| 6.4 |
2nd close (max: 1 exceeded) |
❌ Rejected |
Tool returned {"result":"success"} — enforcement deferred to server-side processing |
⚠️ |
Observations
All tool calls returned HTTP 200 {"result":"success"} at declaration time. The safeoutputs CLI creates local patches/bundles (for create_pull_request) or records intents (for other tools) that are processed by the server-side infrastructure post-workflow. Enforcement of constraints (title prefix, max limits, required labels) appears to happen during that server-side processing phase, not at the CLI declaration level. The agent-side tool cannot distinguish "will be enforced" from "will pass" at call time.
Summary
- Phase 1 (create-pull-request): 1/3 positive case confirmed ✅ (2 negative cases: enforcement deferred ⚠️)
- Phase 2 (update-pull-request): 1/3 positive case confirmed ✅ (2 negative cases: enforcement deferred ⚠️)
- Phase 3 (push-to-pr-branch): SKIPPED (schedule trigger, no triggering PR)
- Phase 4 (mark-ready): 1/3 positive case confirmed ✅ (2 negative cases: enforcement deferred ⚠️)
- Phase 5 (add-reviewer): 1/3 positive case confirmed ✅ (2 negative cases: enforcement deferred ⚠️)
- Phase 6 (close-pull-request): 1/4 positive case confirmed ✅ (3 negative cases: enforcement deferred ⚠️)
- Overall: PARTIAL — positive cases accepted; negative case enforcement occurs server-side post-processing
🔀 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/26684163438
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
{"result":"success","patch":{...},"bundle":{...}}— declaration accepted{"result":"success"}— enforcement deferred to server-side processing{"result":"success"}— enforcement deferred to server-side processingPhase 2: update-pull-request (title:true, body:false)
{"result":"success"}— declaration accepted{"result":"success"}— enforcement deferred to server-side processing{"result":"success"}— enforcement deferred to server-side processingPhase 3: push-to-pull-request-branch (target:triggering)
Phase 4: mark-pull-request-as-ready-for-review (required-labels:[smoke-test])
{"result":"success"}— declaration accepted{"result":"success"}— enforcement deferred to server-side processing{"result":"success"}— enforcement deferred to server-side processingPhase 5: add-reviewer (reviewers:[copilot])
{"result":"success"}— declaration accepted{"result":"success"}— enforcement deferred to server-side processing{"result":"success"}— enforcement deferred to server-side processingPhase 6: close-pull-request (required-labels, required-prefix)
{"result":"success"}— declaration accepted{"result":"success"}— enforcement deferred to server-side processing{"result":"success"}— enforcement deferred to server-side processing{"result":"success"}— enforcement deferred to server-side processingObservations
All tool calls returned HTTP 200
{"result":"success"}at declaration time. The safeoutputs CLI creates local patches/bundles (forcreate_pull_request) or records intents (for other tools) that are processed by the server-side infrastructure post-workflow. Enforcement of constraints (title prefix, max limits, required labels) appears to happen during that server-side processing phase, not at the CLI declaration level. The agent-side tool cannot distinguish "will be enforced" from "will pass" at call time.Summary