Skip to content

[WIP] Fix agent-performance-analyzer to use local GitHub MCP#7207

Closed
Copilot wants to merge 2 commits intomainfrom
copilot/fix-agent-performance-analyzer
Closed

[WIP] Fix agent-performance-analyzer to use local GitHub MCP#7207
Copilot wants to merge 2 commits intomainfrom
copilot/fix-agent-performance-analyzer

Conversation

Copy link
Contributor

Copilot AI commented Dec 22, 2025

Plan: Fix agent-performance-analyzer to use local GitHub MCP

  • Understand the issue - agent-performance-analyzer.md is missing mode: field in GitHub tools configuration
  • Add mode: local to the GitHub tools configuration in agent-performance-analyzer.md
  • Recompile the workflow lock file with make recompile
  • Validate the changes - confirmed lock file now uses Docker-based local MCP server
  • Run agent-finish to ensure everything passes
Original prompt

This section details on the original issue you should resolve

<issue_title>[q] Fix agent-performance-analyzer to use local GitHub MCP</issue_title>
<issue_description># 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:

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 githubnext/gh-aw#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

Comments on the Issue (you are @copilot in this section)


✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

Co-authored-by: pelikhan <4175913+pelikhan@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

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

2 participants