autoskill

To extract coding preferences from session history, the skill detects recurring corrections and approvals to identify durable patterns. It then maps these signals to active skill files and proposes minimal, reversible edits for review.

Automating skill improvement from session analysis involves mapping detected corrections to their corresponding skill files. The system proposes targeted, minimal edits to refine active skills, codifying durable preferences for future cross-session and cross-project use.

Yes, you can use pattern detection to apply cross-session learning across different projects where your skills are active. It extracts durable preferences from coding sessions and updates skill configurations to improve future interactions regardless of the project context.

To propose reversible edits to yaml-frontmatter skill files, map session correction signals to the relevant active skills. The system generates minimal configuration changes for your review, allowing you to apply only approved updates and maintain a reversible history.

You need to run code-correction analysis when you want to remember a specific pattern or update your skills after finishing a coding session. It detects recurring feedback to ensure durable preferences are codified into the appropriate skill files.

No dependencies are required to detect repeated feedback through session analysis. The system independently evaluates your coding session history to identify learnable patterns and propose targeted improvements to the active skills used.