Where are the learnings stored?

Project-specific insights are typically stored in .claude/CLAUDE.md or documentation files within your repository, while general patterns can be saved to global skill files.

Does it update my files without permission?

No. The skill follows a strict confirmation workflow where it presents the proposed changes and a diff for your approval before writing any data to disk.

When is the best time to trigger this skill?

You should trigger it after successfully resolving a bug, fixing a build issue, discovering a non-obvious code pattern, or whenever you want to 'save' knowledge for the next session.

What does the Capture Learning skill do?

It analyzes your recent conversation with Claude to identify valuable insights, then formats and saves them into your project's documentation or global skill files.

Can it distinguish between different types of documentation?

Yes, it categorizes learnings into sections like Build & Setup, Troubleshooting, Environment, and Patterns to keep your documentation organized and scannable.

Capture Learning

Name: Capture Learning
Author: petekp

bypetekp

•

Learning & Documentation

Extracts actionable insights from conversation history to update project documentation and persistent agent knowledge.

About

The Capture Learning skill transforms fleeting conversation insights into permanent project intelligence. By analyzing recent debugging sessions, build resolutions, and architectural decisions, it identifies the root causes of issues and documents their solutions. It intelligently distinguishes between project-specific configurations (stored in CLAUDE.md) and general coding patterns (stored as global skills), ensuring that your AI assistant becomes more effective with every interaction. This skill is essential for teams looking to maintain high-quality documentation and for individual developers who want to avoid repeating the same troubleshooting steps across different environments.

Key Features

Contextual analysis of debugging sessions to extract problem-cause-solution patterns
Structured templates for build issues, environment setup, and coding conventions
1 GitHub stars
Automated discovery and updates for .claude/CLAUDE.md and project documentation
Interactive diff previews and confirmation workflow before any file modification
Intelligent scope detection for project-specific vs. cross-project knowledge

Use Cases

Capturing a general framework best practice as a reusable skill for other projects
Recording project-specific naming conventions or architectural patterns during development
Documenting a complex workaround for a build error to prevent future recurrence

About

Key Features

Contextual analysis of debugging sessions to extract problem-cause-solution patterns
Structured templates for build issues, environment setup, and coding conventions
1 GitHub stars
Automated discovery and updates for .claude/CLAUDE.md and project documentation
Interactive diff previews and confirmation workflow before any file modification
Intelligent scope detection for project-specific vs. cross-project knowledge

Use Cases

Capturing a general framework best practice as a reusable skill for other projects
Recording project-specific naming conventions or architectural patterns during development
Documenting a complex workaround for a build error to prevent future recurrence