continuous-learning

You can extract coding patterns from AI sessions by running a post-session hook that analyzes transcripts to identify and store successful debugging techniques. This automated pattern detection requires jq for JSON parsing to evaluate and save workflows locally.

Saving debugging techniques involves using a post-session evaluation script that filters out insignificant sessions and stores high-quality resolutions in a local knowledge base directory. This ensures successful project-specific conventions are captured for future reuse.

Yes, you need jq installed to automate pattern extraction because the skill uses jq for JSON parsing. The shell environment accesses this dependency to evaluate session transcripts and store extracted workflows in a local directory.

Session evaluation filters out short or insignificant AI coding sessions automatically before storing learned patterns. This ensures only high-quality debugging techniques and successful error resolutions are saved to your persistent local knowledge base.

Yes, you can capture project-specific conventions after complex tasks because the skill operates as a post-session hook to analyze transcripts. It automatically extracts the successful refactoring approach and saves it as a learned skill for similar future tasks.

Limitations include dependency on a shell environment and jq for JSON parsing, meaning pattern extraction is confined to local directories. The automated hook also only stores sessions that pass its significance filtering, potentially skipping minor workflows.