Install a code reviewer, debugger, or security auditor agent into Claude Code with a single command.
Add reference skills covering topics like database migrations, secure auth, or API pagination for the assistant to consult.
Chain multiple agents into a gated pipeline that reproduces a bug, writes a test, patches it, and reviews the fix.
Let agents build shared memory about your codebase that your whole team benefits from over time.
| vanara-agents/skills | boneskull/promwrap | botts7/dreame-scheduler-addon | |
|---|---|---|---|
| Stars | 7 | 7 | 7 |
| Language | JavaScript | JavaScript | JavaScript |
| Last pushed | — | 2020-07-21 | — |
| Maintenance | — | Dormant | — |
| Setup difficulty | easy | easy | moderate |
| Complexity | 2/5 | 2/5 | 2/5 |
| Audience | developer | developer | general |
Figures from each repo's GitHub metadata at analysis time.
Runs on an existing Claude subscription with no separate API key required.
This repository is a free collection of prebuilt agents and skills for Claude Code, the AI coding assistant. Rather than writing your own instructions from scratch for common development tasks, you can install ready-made ones with a single command, and they run using your existing Claude subscription rather than requiring a separate metered API key. The project describes itself as the free tier of a larger paid catalog, with this open portion containing around thirty items. The collection is split into agents and skills. Agents behave like specialists that activate for a specific moment in the workflow, such as a code reviewer that runs after code changes, a debugger built around forming and testing hypotheses, a security auditor that checks code touching things like authentication or database queries, and a test author that writes tests before the implementation. Skills are more like reference playbooks the assistant reads while working, covering topics such as API pagination, caching strategies, database migrations, secure authentication, SQL index tuning, and writing clear commit messages following the Conventional Commits format. Each item is packaged as a full directory rather than a single instructions file, including reference documents, worked examples, and runnable verification scripts that are checked automatically in the project's continuous integration on every update. There is also an orchestration skill that chains multiple agents together into a pipeline, for example reproducing a bug, writing a test, applying a fix, reviewing it, and committing, with checks at each stage that stop the process if something fails. Agents can also record what they learn about your codebase into a memory folder that your team can share by committing it. This project suits developers already using Claude Code who want a ready set of specialist agents and reference skills for tasks like code review, security auditing, refactoring, and API design, without building that tooling themselves. It is released under the Apache-2.0 license.
A free, Apache-2.0 collection of ready-made specialist agents and reference skills you can install into Claude Code with one command.
Mainly JavaScript. The stack also includes JavaScript, Claude Code, Node.js.
Use freely for any purpose, including commercial use, as long as you keep the copyright notice.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
double-check against the repo, no cap.