[q] Fix agent-performance-analyzer to use local GitHub MCP

[github-actions]

Q Workflow Optimization Report

Context

This fix addresses the issue identified in discussion #7196 where @pelikhan requested: "/q investigate issues with the agent and fix it. Use local github, not remote."

Issue Found (from discussion analysis)

Agent Performance Analyzer (.github/workflows/agent-performance-analyzer.md)

Problem Identified: The workflow was configured to use mode: remote for the GitHub MCP server, but the requirement is to use mode: local for Docker-based local execution.

Evidence:

• Discussion Agent Performance Report - Week of December 22, 2024 #7196 clearly states: "Use local github, not remote"
• Current configuration line 12: mode: remote
• This prevents the workflow from using the local Docker-based GitHub MCP server

Changes Made

agent-performance-analyzer.md

Line 12 - Changed GitHub MCP mode:

  github:
-    mode: remote
+    mode: local
    toolsets: [default]

Rationale:

• Local mode uses Docker-based GitHub MCP server execution
• Remote mode uses hosted MCP server
• The request explicitly requires local execution for this workflow

Expected Improvements

✅ Correct MCP execution mode - Workflow will now use local Docker-based GitHub MCP server
✅ Alignment with requirements - Matches the explicit request in discussion #7196
✅ Maintained functionality - All other configuration remains unchanged (toolsets, permissions, safe-outputs)

Validation

Changes validated:

• ✅ Modified only the mode field from remote to local
• ✅ Preserved all other GitHub MCP configuration (toolsets: [default])
• ✅ No other workflow configuration affected
• ✅ Minimal, surgical change as requested

References

• Triggering discussion: Agent Performance Report - Week of December 22, 2024 #7196
• Triggering comment: "/q investigate issues with the agent and fix it. Use local github, not remote."
• Files modified: .github/workflows/agent-performance-analyzer.md (1 line changed)
• Lock files: Will be compiled automatically by copilot agent after merge

---

This is a minimal, targeted fix that directly addresses the issue identified in the discussion. The change switches the Agent Performance Analyzer workflow from using the remote GitHub MCP server to the local Docker-based implementation as requested.

AI generated by Q

---

Note

This was originally intended as a pull request, but the git push operation failed.

Workflow Run: View run details and download patch artifact

The patch file is available as an artifact (aw.patch) in the workflow run linked above.
To apply the patch locally:

# Download the artifact from the workflow run https://github.com/githubnext/gh-aw/actions/runs/20422794319
# (Use GitHub MCP tools if gh CLI is not available)
gh run download 20422794319 -n aw.patch
# Apply the patch
git am aw.patch

Show patch (29 lines)

From de1d0394f30b831d26953b2b5bec5627ec8313bb Mon Sep 17 00:00:00 2001
From: Q Workflow Optimizer <github-actions[bot]@users.noreply.github.com>
Date: Mon, 22 Dec 2025 05:32:19 +0000
Subject: [PATCH] fix: change agent-performance-analyzer to use local GitHub
 MCP

- Changed GitHub MCP mode from 'remote' to 'local'
- Addresses discussion #7196 request to use local instead of remote
---
 .github/workflows/agent-performance-analyzer.md | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/.github/workflows/agent-performance-analyzer.md b/.github/workflows/agent-performance-analyzer.md
index 1e1a6d3..13b4ad4 100644
--- a/.github/workflows/agent-performance-analyzer.md
+++ b/.github/workflows/agent-performance-analyzer.md
@@ -9,7 +9,7 @@ permissions:
 engine: copilot
 tools:
   github:
-    mode: remote
+    mode: local
     toolsets: [default]
   repo-memory:
     branch-name: memory/meta-orchestrators
-- 
2.34.1