Automates runtime log collection and hypothesis testing to fix complex bugs using empirical evidence instead of guesswork.
The Runtime Evidence Debugger skill provides a structured, hypothesis-driven workflow for resolving difficult software defects. By automating the setup of a local logging server and providing standardized instrumentation patterns, it allows developers to capture real-time values, state transitions, and execution flows. This skill is especially useful for debugging issues that are hard to reproduce or require specific user interactions, as it replaces manual DevTools inspection with programmatically accessible, structured logs that Claude can analyze directly to confirm or reject root-cause hypotheses.
주요 기능
01Automated local log server setup with session-based tracking
02Hypothesis-driven debugging framework to validate specific root causes
03Multi-language instrumentation snippets for JavaScript, TypeScript, and Python
04Structured NDJSON log analysis for clear evidence-based conclusions
059,883 GitHub stars
06Pre- and post-fix verification workflow to ensure permanent resolution
사용 사례
01Debugging complex state issues or null/undefined errors in frontend applications
02Identifying async timing issues and race conditions in distributed systems
03Capturing runtime evidence for bugs that depend on specific user interactions
