Skill UI

This internal agent skill defines a systematic, multi-step procedure for scanning GitHub repositories to gather comprehensive evidence supporting or explaining bug clusters. The process covers searching open issues, inspecting recent commits, analyzing code paths, and checking deployment history. It assigns evidence confidence tiers (Exact, Strong, Moderate, Weak) to provide a deep, actionable understanding of potential software defects.

This skill provides comprehensive code review capabilities for analyzing git diffs or specific file scopes. It goes beyond simple syntax checks, assessing code for reuse opportunities, quality issues (e.g., redundant state, leaky abstractions), efficiency bottlenecks, and adherence to standards. It utilizes multiple specialized sub-agents (Reuse, Quality, Efficiency, Clarity) to perform parallel, read-only audits. Optionally, it can apply high-confidence, behavior-preserving fixes upon user request.

A comprehensive 7-section readiness report generated for a repository, analyzing key metrics including risk score, test confidence, and publish readiness. It assesses architectural viability, agent topology recommendations, and potential risks, making it essential for pre-mint reviews, CI/CD gating, and detecting architectural drift in large codebases.

This skill executes the metaharness score to provide a comprehensive, multi-dimensional readiness scorecard for a codebase. It assesses critical dimensions like compilation confidence, task coverage, tool safety, and memory usefulness. Use it to evaluate if a repository is prepared for harness adoption before initiating development tools or framework setup.

Compute weighted similarity between two harness fingerprints using cosine, categorical agreement, and topology Jaccard. Use it to rank templates, detect drift, decide reuse vs scaffold, and generate confidence scores for recommender workflows.

Executes a multi-agent, structured code review workflow for GitHub Pull Requests (PRs). It analyzes changes from multiple perspectives (e.g., bug scans, compliance with project guidance, git history, etc.), provides confidence scoring, and generates actionable, highly formatted feedback directly on the PR. Use this when a thorough, professional code review is required.