How does the Continuous Learning skill trigger?

It functions as a 'Stop' hook that executes automatically at the end of a Claude Code session, ensuring it captures the full context without interrupting your coding flow.

Where are the learned skills stored?

By default, all identified patterns and skills are saved to the ~/.claude/skills/learned/ directory for easy review and management.

Can I prevent the skill from saving minor typos or one-off fixes?

Yes, the skill includes a config.json file where you can define 'ignore_patterns' to filter out simple typos, one-time fixes, or external API issues.

Does this skill add latency to my AI messages?

No, it is designed to be lightweight and non-blocking, only running once a session is completed so there is zero impact on message response times.

How is this different from the standard /learn command?

While the /learn command requires manual user intervention during a session, Continuous Learning automatically discovers and suggests skills based on session history.

Continuous Learning Agent

Name: Continuous Learning Agent
Author: affaan-m

byaffaan-m

•

112,913

•

Productivity & Workflow

Extracts reusable coding patterns and troubleshooting techniques from AI sessions to build a persistent, self-evolving knowledge base.

The Continuous Learning skill optimizes developer workflows by automatically analyzing completed Claude Code sessions to identify repeatable patterns, error resolutions, and project-specific workarounds. By operating as a non-blocking stop hook, it evaluates session context and message history to generate new, localized skills without interrupting active development. This ensures that once a complex problem is solved, the solution is captured as a permanent asset in the user's skill directory for future automation.

Key Features

01Automated session evaluation via customizable stop hooks

02Persistent local storage of learned skills for future session reuse

03Support for project-specific architectural conventions and debugging techniques

04112,913 GitHub stars

05Intelligent pattern detection for error resolutions and workarounds

06Configurable extraction thresholds for session length and message volume

Use Cases

01Automatically documenting unique framework bug fixes for future reference

02Capturing specific project coding standards without manual documentation

03Scaling AI knowledge by converting historical session data into reusable developer skills

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add affaan-m/everything-claude-code continuous-learning

For use in Claude.ai and ChatGPT

Key Features

01Automated session evaluation via customizable stop hooks

02Persistent local storage of learned skills for future session reuse

03Support for project-specific architectural conventions and debugging techniques

04112,913 GitHub stars

05Intelligent pattern detection for error resolutions and workarounds

06Configurable extraction thresholds for session length and message volume

Use Cases

01Automatically documenting unique framework bug fixes for future reference

02Capturing specific project coding standards without manual documentation

03Scaling AI knowledge by converting historical session data into reusable developer skills

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add affaan-m/everything-claude-code continuous-learning

For use in Claude.ai and ChatGPT