engineering / skill

Claude Code Skills & Plugins

Agent skills, plugins and personas for engineering, product, marketing, compliance, research and business work in Claude Code, Codex, Gemini CLI and more.

Repository owner alirezarezvani

Not rated yet
0 comments

Like 0 Save

Published 11 Oct 2026Listing updated 12 Oct 2026

At a glance

Purpose: Agent skills, plugins and personas for engineering, product, marketing, compliance, research and business work in Claude Code, Codex, Gemini CLI and more.
Runs in: Local agent with shell
Operating systems: Any
Needs: Runtime: Python; Needs internet: Yes; Needs accounts or API keys beyond your agent: Not stated
Files and data: Reads files: Yes; Writes files: Yes; Sends data to other services: Not stated
Tool type: Skill
Delivery: From GitHub
Version: Latest on the main branch
Access: Code on GitHub
Compatibility: According to the README: Claude Code, OpenAI Codex, Gemini CLI, OpenClaw, Hermes Agent, Mistral Vibe, Cursor, Aider, Windsurf, Kilo Code, OpenCode, Augment and Antigravity. Hermes Agent and Mistral Vibe need a one-time local sync. The README says the Python tools run anywhere Python runs.
External costs: The repository is MIT-licensed and the README states no pricing. Your coding agent or model provider may charge for usage.
Permissions: The skills act through your coding agent with the access you give it. The README describes Python command-line tools and security PreToolUse hooks. The README describes skillopt-sleep as nightly gated self-evolution from real Claude Code sessions, and agent-memory as a memory workflow over Claude Code hooks, with CLAUDE.md changes through human adoption. The README lists worktree lifecycle and cleanup automation scripts. The README's OpenClaw route executes a downloaded shell installer with Bash. The README describes Agent Launcher workflows to build, launch and schedule Claude Managed Agents in your own account with your own API keys, including a cron deployment loop. The README shows skill and persona installation into ~/.claude/skills/, ~/.claude/agents/, ~/.codex/skills/, ~/.hermes/skills/, ~/.vibe/skills/ and ~/.gemini/antigravity/skills/, plus project-level rule and skill files for other hosts.
Usage terms: MIT License, as stated in the repository. See its LICENSE file for the exact terms. The README describes Agent Launcher as a reimplementation of Anthropic's Apache-2.0 launch-your-agent and identifies vendored components from other projects, without stating separate licence terms for those components.

Code on GitHub

alirezarezvani/claude-skills

MIT

Python28,075 stars3,933 forksLast pushed 30 Aug 2026 UTC

380 Claude Code skills & agent skills & plugins (30+ Agents, 70+ custom commands, 380+ skills, customizable references, scripts)for Claude Code, Codex, Gemini CLI, Cursor, and 8 more coding agents — engineering, marketing, product, compliance, C-level advisory, research, business operations, commerc

Topics: agent-plugins agent-skills agentic-ai ai-coding-agent anthropic-claude claude-ai claude-code claude-code-plugins

Public GitHub data fetched 13 Oct 2026 UTC · 0 days old

View on GitHub ↗

What it does

This repository is a library of agent skills, plugins, agents and personas. The README describes SKILL.md instructions, Python tools and reference documents, grouped into domains including engineering, product, marketing, regulatory and quality, compliance, C-level advisory, finance, business operations, commercial work, research and productivity.

The README names Claude Code, OpenAI Codex, Gemini CLI, OpenClaw, Hermes Agent, Mistral Vibe, Cursor, Aider, Windsurf, Kilo Code, OpenCode, Augment and Antigravity. The README provides a Claude Code plugin marketplace, native skill and sync routes for several hosts, and conversion and installation scripts for other tools.

The README lists Startup CTO, Growth Marketer and Solo Founder personas and describes orchestration patterns combining skills, agents and personas. The README describes its skill-security-auditor feature as scanning skill folders for patterns including command injection and data exfiltration and returning PASS, WARN or FAIL with remediation guidance.

Who it is for

People who use coding agents such as Claude Code or Codex and want ready-made skills and personas for engineering, marketing, product, compliance and business roles.

Example tasks

Review a microservices architecture for scalability risks with the senior-architect skill
Write an SEO-optimized blog post with the content-creator skill
Check technical documentation for MDR Annex II compliance gaps with the mdr-745-specialist skill
Follow the README's skill-security-auditor example to scan a skill folder before installation
Prioritize features from a CSV file with the RICE prioritizer script

Getting started

The README gives a route per tool. For Claude Code, run /plugin marketplace add alirezarezvani/claude-skills and then install a bundle such as /plugin install engineering-skills@claude-code-skills. For Codex, run npx agent-skills-cli add alirezarezvani/claude-skills --agent codex. For Gemini CLI, clone the repository and run ./scripts/gemini-install.sh. For other tools, it describes ./scripts/convert.sh --tool all, followed by ./scripts/install.sh --tool cursor --target . for Cursor. You can also copy a skill folder into ~/.claude/skills/ or ~/.codex/skills/. The README has the exact commands.

Limitations

According to the README, you use one of the supported coding agents, and install steps differ by tool: a plugin marketplace for Claude Code, scripts for Gemini CLI and Mistral Vibe, a sync script for Hermes Agent, and converted formats for Cursor, Aider, Windsurf and others. It tells Windows users to clone with git clone -c core.symlinks=true and Developer Mode enabled, and to set PYTHONUTF8=1. Many skills include Python tools. The README says they use only the standard library, with optional third-party parsers for some document extractors. The repository can change at any time; check its README before you install.

Data practices

The README gives no overall statement on data handling. The README describes standard-library Python tools and optional third-party parsers for book-to-skill document extraction; it does not explain those parsers' data handling. It names some skills that reach outside services: loop-library reads a live catalog from an outside site at runtime, and the project-management skills bundle the Atlassian Remote MCP. What is sent where depends on the skills you use, your coding agent and your model provider.