Transforms Claude Code chat histories into structured, reusable SKILL.md files using Fabric AI patterns.
Transform valuable but messy Claude Code conversations into production-ready skill files by distilling trial-and-error into actionable workflows. This skill automates the extraction of wisdom, step-by-step instructions, and core problem-solution patterns from JSONL session logs using specialized Fabric AI patterns. By filtering out dead ends and noise, it allows developers to build a robust library of reusable tribal knowledge and proven implementation recipes from their past development sessions, ensuring that hard-won solutions are never lost.
Key Features
01Standardized SKILL.md template generation for consistent documentation
02Noise filtering to isolate the successful 'critical path' from trial-and-error
03Automated parsing of Claude Code JSONL session logs into readable text
041 GitHub stars
05Parallel processing pipeline for rapid distillation of conversation data
06Multi-pattern extraction of insights, instructions, and problem-solution pairs
Use Cases
01Capturing implementation workflows from research and learning conversations
02Converting a complex debugging session into a reusable troubleshooting skill
03Building a library of project-specific best practices from successful sessions
