What problem does it solve?
This Skill helps you turn the friction from a just-finished work session into concrete improvements instead of letting useful lessons disappear. It is designed for cases where work shipped successfully, but the session still exposed preventable inefficiencies like tool retries, repeated edits, manual verification steps, or missing guidance.
Core Features & Use Cases
- Friction-based improvement detection: Scans recent work for grounded signals such as blocked tool calls, redo edits, unnecessary confirmation loops, and manual cross-referencing.
- Dedup and routing: Checks whether the pattern is already covered in memories, workflow notes, rules, or skills, then routes the improvement to the right durable surface.
- Build-first workflow: Proposes a specific edit to a hook, skill, memory file, or related artifact, with explicit user approval before anything is written.
- Fallback capture: If no concrete buildable fix exists, it records the softer pattern as habit memory so it can be promoted later if it repeats.
- Use case: After a PR merge or a completed investigation, use this Skill to identify what slowed the session down and convert that into a reusable automation, rule, or skill enhancement.
Quick Start
Ask the AI to run the compound skill after a completed work unit and identify buildable improvements from the session friction.