skillhub
The skill registry for AI coding agents.
skillhub install security-review
skillhub install pipe --platform cursor
skillhub search refactor
npm for AI coding agent skills. One command to find, install, and publish.
The Problem
AI coding agent skills spread via blog posts, Substack threads, and copy-paste.
There is no npm install. No version control. No discovery. No standard format.
You either write skills from scratch or spend 30 minutes hunting through GitHub for one someone else built.
The Solution
skillhub is a registry for AI coding agent skills — like npm, but for SKILL.md files.
pip install skillhub
skillhub search security # find what you need
skillhub install security-review # install in one command
skillhub publish # share what you build
Skills install to the right place automatically based on your agent. No manual copy-paste.
Install
pip install skillhub
Usage
Search
skillhub search security
skillhub search refactor
skillhub search pipeline
Install
skillhub install pipe # auto-detects your agent
skillhub install graphify --platform claude # specific platform
skillhub install security-review --platform cursor
List all skills
skillhub list
Get skill details
skillhub info pipe
Publish your skill
skillhub publish
Supported Agents
| Agent | Install location |
|---|---|
| Claude Code | ~/.claude/skills/<name>/SKILL.md |
| Codex | appended to AGENTS.md |
| Cursor | .cursor/rules/<name>.mdc |
| Aider | appended to AGENTS.md |
| OpenCode | appended to AGENTS.md |
| Claw | appended to AGENTS.md |
| Gemini CLI | appended to GEMINI.md |
skillhub auto-detects your agent. Pass --platform to override.
Registry
Skills in the registry:
| Skill | Description | Author |
|---|---|---|
pipe |
Chain skills into pipelines with -> syntax |
@aayuvraj |
graphify |
Turn any codebase into a queryable knowledge graph | @safishamsi |
review |
Pre-landing PR review — SQL, trust boundaries, side effects | @garrytan |
investigate |
Root cause debugging — 4-phase systematic process | @garrytan |
code-review |
Multi-axis review: correctness, security, performance | @addyosmani |
debug |
Structured triage for failures, errors, and incidents | @addyosmani |
tdd |
Test-driven development — failing test first, then code | @addyosmani |
plan |
Decompose work into verifiable tasks with acceptance criteria | @addyosmani |
spec |
Full specification before any code — requirements, scope, criteria | @addyosmani |
performance |
Profiling-driven optimization — Core Web Vitals, load times | @addyosmani |
harden |
Security-first practices — auth, data storage, third-party inputs | @addyosmani |
document |
ADRs and docs capturing the why, not just the what | @addyosmani |
git |
Disciplined git — commits as save points, history as docs | @addyosmani |
changelog |
Generate release notes from Conventional Commits | @alirezarezvani |
security-review |
Deep security audit — OWASP, injection, auth flaws | community |
explain |
Plain-language code explanation with edge cases | community |
refactor |
Refactor for clarity and performance, behavior unchanged | community |
Registry grows via community PRs. See Contributing.
Publishing a Skill
Built something useful? Add it to the registry.
- Fork this repo
- Add
registry/skills/<your-skill>.md - Add an entry to
registry/index.json - Open a PR
Skill format:
---
name: your-skill
description: one line description
trigger: /your-skill
platforms: claude, codex, cursor, aider, claw, opencode, gemini
---
# /your-skill
Instructions for the agent...
Philosophy
Skills are the most underrated part of AI-assisted development.
A great skill turns a 5-minute back-and-forth into one command. A great skill library turns any agent into a specialist. But right now, skills are scattered, undiscoverable, and agent-specific.
skillhub is the missing layer: a standard format, a central registry, and a one-command install that works across every major agent.
Build once. Share everywhere.
Contributing
- Add a skill to the registry
- Improve an existing skill
- Add platform support for a new agent
PRs welcome. See Publishing a Skill.
License
MIT
If this saved you time, a ⭐ helps others find it.
Metadata
Release files for skillhub-cli 0.1.0
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | |
|---|---|---|---|---|
| skillhub_cli-0.1.0.tar.gz | 5.9 kB | Without provenance | Details |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Reset |
|---|---|---|---|---|
| skillhub_cli-0.1.0-py3-none-any.whl | Python 3 | none | any | Details |
Total release size: 12.0 kB
Release files / skillhub_cli-0.1.0.tar.gz
| Download URL | skillhub_cli-0.1.0.tar.gz |
|---|---|
| Size | 5.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
9118eff6148442c1aaa3811c7588bf4e35c53ce260653bafcad6afeaf2396d04
|
|
BLAKE2b-256 checksum How to use checksums |
0fa44595aeca1a74716af4133682dd4bb91cd900aa95c0f4cf3165f492704606
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.3
|
Release files / skillhub_cli-0.1.0-py3-none-any.whl
| Download URL | skillhub_cli-0.1.0-py3-none-any.whl |
|---|---|
| Size | 6.1 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
194df0c863dd4d2493efc618e0b90aefee66e914b418bb6c10b996a9e350cadb
|
|
BLAKE2b-256 checksum How to use checksums |
acfd7d77c48c66135f1d54eaf018ac6f8e9641466fe07b757941a90782af4d34
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/6.2.0 CPython/3.14.3
|
Security for skillhub-cli 0.1.0
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.