GitHub research tools for GAIA benchmark evaluation - issue timeline, label history, advanced search
- Copy this configuration to your Claude Desktop config file:
{
"mcpServers": {
"mcp-github-research": {
"command": "node",
"args": ["/Users/bard/Code/mcp-github-research/src/index.js"]
}
}
}-
Restart Claude Desktop
-
Test: Ask Claude "List GitHub research tools" or "Search for numpy regression issues"
- github_search_issues - Search issues across repositories with advanced filtering
- github_get_issue_timeline - Get complete timeline including label changes (perfect for GAIA!)
- github_get_issue_events - Get specific events for an issue
- github_list_repository_issues - List issues from a repository with filtering
- github_analyze_label_history - Analyze when labels were added (solves GAIA Question 3!)
These tools specifically solve the GitHub timeline/label research problems in GAIA:
- Question Type: "When was X label added to the oldest Y issue?"
- Solution:
github_analyze_label_history+github_get_issue_timeline - Example: "According to github, when was Regression added to the oldest closed numpy.polynomial issue that has the Regression label?"
User: "When was the regression label added to the oldest numpy polynomial issue?"
Claude: I'll search for closed numpy polynomial issues with the regression label and analyze when the label was added.
[Uses github_analyze_label_history with owner="numpy", repo="numpy", label_name="regression"]
Result: The oldest issue was #22669, regression label added on 11/25/22
For higher rate limits, add a GitHub token:
{
"mcpServers": {
"mcp-github-research": {
"command": "node",
"args": ["/Users/bard/Code/mcp-github-research/src/index.js"],
"env": {
"GITHUB_TOKEN": "your_token_here"
}
}
}
}- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%/Claude/claude_desktop_config.json - Linux:
~/.config/Claude/claude_desktop_config.json
- ✅ Issue search with complex queries
- ✅ Timeline analysis with label events
- ✅ Date formatting (MM/DD/YY)
- ✅ Chronological sorting
- ✅ Rate limiting for API calls
- ✅ Error handling
- ✅ Works without authentication (public repos)
This tool specifically addresses the GitHub research limitations identified in GAIA Question 3 and similar benchmark questions requiring GitHub timeline analysis.