EfanWang

EfanWang / skills-manager

Public

Manage, install, update, and trace agent skills (Claude Code, Cursor, Codex CLI)

30
2
89% credibility
Found May 29, 2026 at 30 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

A skill management tool that helps users organize, track, and update AI assistant skills by recording where each skill originated and checking for new versions from GitHub.

How It Works

1
💡 Discovering the need for organization

You use AI coding assistants and have collected many skills over time, but you can't remember where each one came from.

2
📦 Installing the manager

You place the skills-manager folder alongside your other skills, and it immediately starts working—no setup required.

3
🔍 Scanning your skill library

You run a quick scan and see every skill you have installed, complete with descriptions and where each one originated.

4
Finding out which skills need attention
⬆️
Skills with updates available

You choose to pull the latest versions and your skills are refreshed automatically with a backup kept just in case

🔎
Unidentified skills

The manager helps you trace where unknown skills came from by checking your files and online sources

✨ Your skills are organized and current

Every skill is tracked, up-to-date, and you know exactly where each one came from—no more guessing or lost sources.

Sign up to see the full architecture

3 more

Sign Up Free

Star Growth

See how this repo grew from 30 to 30 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is skills-manager?

skills-manager is a Python-based tool that helps you manage, update, and track AI agent skills across Claude Code, Cursor, and Codex CLI. It solves the messy problem of losing track of where your skills came from, whether they have updates available, and what happens if you accidentally overwrite local changes. The tool maintains a lightweight registry file that records the GitHub source for each skill, then provides CLI scripts to scan your skills directory, check for upstream updates, install new skills from GitHub URLs, and update existing ones with automatic backups before any changes.

Why is it gaining traction?

The SKILL.md ecosystem is growing fast, but nobody built the boring infrastructure layer yet. Once you install five or ten skills from different GitHub repos, you start asking questions like "did I modify this locally?" and "is there a newer version upstream?" This tool answers those questions automatically. The source tracing feature is particularly clever: it can identify unknown skills by inspecting their git history or embedded GitHub URLs, then hand off unresolved cases to the agent for web search. Safety-conscious developers appreciate the no-telemetry approach, atomic directory swaps, and automatic backup system that protects against broken updates.

Who should use this?

Power users running multiple skills from different sources who want to stay current without manually tracking each repository. Agents that need to manage their own skill lifecycle programmatically. Teams standardizing on skills who want a clear audit trail of where each skill originated. Developers who frequently experiment with new skills and need a safe way to roll back after updates.

Verdict

This is a genuinely useful tool for a real problem, and the implementation shows careful thinking around safety and edge cases. However, with only 30 stars and a credibility score of 0.8999999761581421%, the project is early-stage and the community is small. Documentation is solid for the core workflows, but test coverage is minimal. Worth installing if you manage multiple agent skills seriously, but treat it as a personal utility rather than production infrastructure until it matures.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.