Generates evidence-based performance hypotheses by analyzing git history and specific execution paths.
The Performance Hypothesis Gatherer is a diagnostic tool for Claude Code designed to identify the root causes of performance regressions. By systematically reviewing recent git history and mapping relevant code paths, the skill produces up to five data-backed hypotheses. Each hypothesis includes specific evidence from the repository and a confidence rating, allowing developers to understand why performance issues are occurring before attempting to optimize. This skill enforces a strict separation between diagnosis and implementation to ensure rigorous debugging.
Key Features
01Evidence-linked hypothesis generation with confidence scoring
02311 GitHub stars
03Automated git history analysis for performance regression tracking
04Targeted code path identification via repo-mapping and grep
05Strict diagnostic constraints to prevent premature optimization
06Standardized YAML-style output for easy integration into dev workflows
Use Cases
01Identifying the specific commit responsible for a recent latency spike
02Mapping complex execution paths to find hidden bottlenecks in legacy code
03Creating a data-driven debugging roadmap for reported performance issues
