How does it know which files are relevant?

The skill analyzes your recent session activity, including git diffs and tool usage, to automatically infer which files relate to the knowledge being documented.

What is a Knowledge Card?

A Knowledge Card is a structured markdown file that captures specific technical insights, such as architectural decisions, constraints, or 'gotchas', to be shared and versioned with your team.

Where are the cards stored?

Cards are saved locally in the .claude/memory/cards/ directory, allowing them to be committed to your repository and shared via Git.

How does the secret scanning work?

Before saving any card, the skill automatically scans the content for patterns like AWS keys, API tokens, and database connection strings, masking them with a redaction warning if found.

Can I create private notes that aren't shared?

Yes, by using the --visibility private flag, you can create cards that are intended for your local use only and will be flagged as not for team distribution.

Knowledge Card Generator

Name: Knowledge Card Generator
Author: TeamSPWK

byTeamSPWK

0•

协作工具

Captures and organizes technical insights into standardized markdown cards for team-wide collective memory.

This skill enables developers to build a persistent collective memory for their team by documenting key technical decisions, architectural constraints, and operational 'gotchas' directly within Claude Code. It automatically gathers context from recent file edits and session history, performs integrated secret scanning to prevent credential leaks, and generates structured markdown files with rich YAML metadata. By saving these insights to a shared directory, it transforms fleeting session context into permanent, version-controlled documentation that helps prevent recurring errors and streamlines onboarding.

主要功能

01Multi-type categorization including decisions, constraints, and runbooks

02Automatic tag inference and file relationship mapping

030 GitHub stars

04Automated context extraction from Git diffs and session history

05Standardized markdown generation with rich YAML frontmatter

06Built-in secret scanning for API keys, tokens, and database URLs

使用场景

01Creating quick-start runbooks and onboarding guides during the development process

02Documenting architectural decisions and their underlying rationale for future reference

03Recording 'gotchas' and debugging solutions to prevent team members from repeating errors

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add teamspwk/claude-session-memory card

For use in Claude.ai and ChatGPT

主要功能

01Multi-type categorization including decisions, constraints, and runbooks

02Automatic tag inference and file relationship mapping

030 GitHub stars

04Automated context extraction from Git diffs and session history

05Standardized markdown generation with rich YAML frontmatter

06Built-in secret scanning for API keys, tokens, and database URLs

使用场景

01Creating quick-start runbooks and onboarding guides during the development process

02Documenting architectural decisions and their underlying rationale for future reference

03Recording 'gotchas' and debugging solutions to prevent team members from repeating errors

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add teamspwk/claude-session-memory card

For use in Claude.ai and ChatGPT