How does this skill help future AI sessions?

It stores insights in a structured LEARNINGS.md file which the ctx system uses to rehydrate Claude's context, preventing the AI from repeating past errors.

What makes a 'learning' worth recording with this skill?

A learning is worth recording if it is specific to your codebase, took real effort to discover, and cannot be easily found via a standard web search.

Should I use this for general code snippets?

No, this skill is designed to capture principles and heuristics rather than raw code snippets or generic programming knowledge available elsewhere.

What structure do the captured learnings follow?

Each entry includes the context of discovery, the core lesson learned, and actionable guidance on how to apply that lesson in the future.

Does this skill require the ctx CLI?

Yes, this skill interacts with the 'ctx' lightweight file-based system via Bash commands to manage project context.

Context Learning Capture

Name: Context Learning Capture
Author: ActiveMemory

byActiveMemory

•

生産性とワークフロー

Records project-specific insights and hard-won lessons to preserve institutional knowledge within your codebase.

Context Learning Capture is a specialized skill for Claude Code that enables the systematic recording of non-obvious project insights, debugging 'gotchas,' and architectural heuristics. By integrating with the 'ctx' file-based context system, it ensures that critical knowledge gained during development sessions is persisted in a structured format (LEARNINGS.md). This skill proactively filters for high-value information—specifically avoiding generic programming knowledge—to provide future AI sessions and human developers with a 'memory' of past challenges and their specific solutions, ensuring that hard-won knowledge is never lost.

主な機能

0136 GitHub stars

02Standardized storage in LEARNINGS.md for easy context rehydration

03Automated recording of debugging 'gotchas' and architectural lessons

04Interactive guidance to refine context, lessons, and applications

05Persistence of project context across multiple AI coding sessions

06Selective capture of non-obvious project insights and heuristics

ユースケース

01Documenting a complex bug root cause that isn't immediately obvious from the code

02Capturing library-specific quirks that required extensive experimentation to understand

03Recording workflow insights to prevent team members from repeating past mistakes

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add activememory/ctx ctx-add-learning

For use in Claude.ai and ChatGPT

Download Skill