--- name: github-issue-analyzer description: Fetches issues from any GitHub repository with recent activity and generates a prioritized report by analyzing issue state, labels, and comment counts to identify high-priority bugs, enhancements, and good first issues for contributors. --- # GitHub Issue Analyzer ## Instructions This skill fetches GitHub issues from any repository and generates a comprehensive prioritized report. It's useful for understanding project health, identifying what needs attention, and helping new contributors find good starting issues. ### Usage 1. Import the analyzer function 2. Call it with a repository owner and name 3. Optionally specify the number of days for "recent activity" (default: 30) 4. The function returns a structured report and saves raw data to `./workspace/` ### Features - **Automatic Categorization**: Issues are categorized into HIGH, MEDIUM, LOW priority and "Good First Issues" - **Smart Filtering**: Identifies bugs vs enhancements, counts active discussions - **Activity-Based Ranking**: Sorts by comment count and recency - **Comprehensive Report**: Includes summary statistics, top 10 recommended issues, and full issue listings ### Priority Logic - **HIGH PRIORITY**: Open bugs OR open issues with >3 comments - **MEDIUM PRIORITY**: Open enhancements/feature requests - **LOW PRIORITY**: Closed issues or less active items - **GOOD FIRST ISSUES**: Issues explicitly labeled as beginner-friendly ## Examples ```typescript import { analyzeGitHubIssues } from './.claude/skills/github-issue-analyzer/implementation'; // Basic usage - analyze last 30 days const report = await analyzeGitHubIssues('modelcontextprotocol', 'inspector'); // Custom timeframe - analyze last 60 days const report = await analyzeGitHubIssues('modelcontextprotocol', 'inspector', 60); // Report includes: // - summary statistics // - top 10 recommended issues with explanations // - categorized issues by priority // - actionable insights and next steps ``` ## Output Files The skill automatically saves: - `./workspace/all-issues.json` - Raw issue data (100 issues per page) - `./workspace/issue-report.md` - Formatted markdown report ## Dependencies - MCP Tool: `list_issues` from GitHub server