[Agent Perf] Deprecation PR #48730 for Matt Pocock Skills Reviewer / Design Decision Gate was closed unmerged — chronic churn co [Content truncated due to length]

[github-actions]

Description

Prior agent-performance memory (agent-performance-latest.md, shared-alerts.md) treated PR #48730 ("Deprecate auto-run paths for Matt Pocock Skills Reviewer and Design Decision Gate") as the resolution to chronic-failure issue #48671. Verified via gh pr view 48730 --json state,mergedAt this run: state: CLOSED, mergedAt: null — the PR was closed without merging, not landed.

Confirmed both workflows still have live on.pull_request triggers in current .github/workflows/mattpocock-skills-reviewer.md and .github/workflows/design-decision-gate.md, and neither sets safe-outputs.report-failure-as-issue: false.

Evidence of continued churn since PR was closed (2026-07-28)

• Matt Pocock Skills Reviewer: 7 new [aw] ... failed issues (e.g. [aw] Matt Pocock Skills Reviewer failed #48774, [aw] Matt Pocock Skills Reviewer failed #48844, [aw] Matt Pocock Skills Reviewer failed #49004, [aw] Matt Pocock Skills Reviewer failed #49121, [aw] Matt Pocock Skills Reviewer failed #49270, [aw] Matt Pocock Skills Reviewer hit AI credits rate limit #49382, [aw] Matt Pocock Skills Reviewer failed #49405)
• Design Decision Gate: 5 new failure issues, including two new distinct failure modes not present in the original churn report:
  • [aw-failures] Design Decision Gate: 'Pre-fetch ADR gate PR context' step fails before agent starts #49144 — 'Pre-fetch ADR gate PR context' step fails before agent starts
  • [aw-failures] Design Decision Gate exhausts LLM invocation cap after permission-denied retries on shell-expanded file paths #49096 — exhausts LLM invocation cap after permission-denied retries on shell-expanded file paths
• Impeccable Skills Reviewer (same original cluster): CLI-hang fix did land (via fix: reduce post-completion idle watchdog and add cleanup timeouts to prevent Copilot CLI hang on exit #44254/fix: add post-result watchdog to codex_harness to prevent CLI hang-on-exit #48499), but hit a new failure mode [aw] Impeccable Skills Reviewer hit AI credits rate limit #49380 (AI credits rate limit) on 2026-07-31.

Expected Impact

Either land the deprecation change (remove on.pull_request, set report-failure-as-issue: false) or make an explicit decision to invest in root-causing the new Design Decision Gate failure modes (#49096, #49144) which are unrelated to the original churn and won't be fixed by deprecation alone. Stops ~12 low-value auto-filed issues/week and unblocks the fleet-wide quality-score plateau (63/100 since Jul 8).

Suggested Agent

Maintainer decision + Copilot SWE Agent to re-submit/land the deprecation change, or a follow-up fix for #49096/#49144 if autonomous execution is to be retained.

Estimated Effort

Low (re-landing #48730 content) to Medium (if root-causing #49096/#49144 is also in scope).

Data Source

Agent Performance Analyzer run 2026-08-01, cross-checked via gh pr view 48730 and gh api search/issues against live .md workflow frontmatter.

Generated by ⚡ Agent Performance Analyzer - Meta-Orchestrator · auto · 53.5 AIC · ⌖ 15.1 AIC · ⊞ 17K · ◷

• expires on Aug 3, 2026, 5:06 AM UTC-08:00

[github-actions]

🍪 Issue Monster selected this for Copilot

I've identified this issue as a good candidate for automated resolution and requested assignment to the Copilot coding agent.

If assignment succeeds, the Copilot coding agent will analyze the issue and create a pull request with the fix.

Om nom nom! 🍪

🍪 Om nom nom by Issue Monster · gpt54 · 3.7 AIC · ⌖ 7.31 AIC · ⊞ 10.5K · ◷

[github-actions]

🍪 Issue Monster selected this for Copilot

I have identified this issue as a good candidate for automated resolution and requested assignment to the Copilot coding agent.

If assignment succeeds, the Copilot coding agent will analyze the issue and create a pull request with the fix.

Om nom nom! 🍪

🍪 Om nom nom by Issue Monster · gpt54 · 3.41 AIC · ⌖ 6.79 AIC · ⊞ 10.7K · ◷

[github-actions]

🍪 Issue Monster selected this for Copilot

I have identified this issue as a good candidate for automated resolution and requested assignment to the Copilot coding agent.

If assignment succeeds, the Copilot coding agent will analyze the issue and create a pull request with the fix.

Om nom nom! 🍪

🍪 Om nom nom by Issue Monster · gpt54 · 5.09 AIC · ⌖ 5.96 AIC · ⊞ 10K · ◷

[github-actions]

🍪 Issue Monster selected this for Copilot

I have identified this issue as a good candidate for automated resolution and requested assignment to the Copilot coding agent.

If assignment succeeds, the Copilot coding agent will analyze the issue and create a pull request with the fix.

Om nom nom! 🍪

🍪 Om nom nom by Issue Monster · gpt54 · 5.49 AIC · ⌖ 6.21 AIC · ⊞ 10K · ◷

[github-actions]

🍪 Issue Monster selected this for Copilot

I identified this issue as a good candidate for automated resolution and requested assignment to the Copilot coding agent.

If assignment succeeds, the Copilot coding agent will analyze the issue and create a pull request with the fix.

Om nom nom! 🍪

🍪 Om nom nom by Issue Monster · gpt54 · 2.9 AIC · ⌖ 6.74 AIC · ⊞ 9.9K · ◷