borghei Claude-Skills
borghei/Claude-SkillsA large library of 385 business, product, compliance and engineering skills with 906 standard-library Python tools, under a Commons Clause license.
With caveats
Check first Depth 4/5
Claude Skills (alirezarezvani) Claude skills
A 400+ skill library of agents, personas and commands for engineering, marketing, product, compliance and C-level work, mirrored for 13 coding agents.
Our verdict Recommended with caveats
Useful as a catalog to pull from, but install per domain: it is broad rather than deep, and some skills overlap with each other and with dedicated packs.
Depth expertise shipped Security Review before installing Maintenance Updated today Adoption stars License MIT, commercial OK
The author changed 2 of 200 skill folders and the plugin manifest or hooks after our review (noticed Oct 29, 2026). The review describes the earlier version; we recheck changed skills weekly, the most significant changes first.
What Claude Skills (alirezarezvani) does
This repository is a catalog rather than a single tool. It bundles roughly 400 skills, over 100 agent definitions, a handful of personas and 150-plus slash commands across engineering, DevOps, product, marketing (including SEO and answer-engine optimization), security, regulatory and compliance, commercial and finance, C-level advisory, productivity, an academic research stack and Markdown-to-HTML document skills. Each skill is a SKILL.md plus, where relevant, reference docs and Python helper scripts; the README claims about 700 stdlib-only scripts and 800 reference files.
You install by domain through the Claude Code plugin marketplace (for example engineering-skills or marketing-skills), or run the Gemini CLI installer, or sync mirrored trees into other agents. The repo ships pre-generated mirror folders for Codex, Gemini CLI, Hermes Agent and Mistral Vibe, and a convert script for tools such as Cursor, Aider, Windsurf and OpenCode. The README separates skills (how to do a task), agents (which task to run) and personas (whose viewpoint to take). Because the scope is so wide, quality varies by domain; the value is in picking the few skills you need rather than loading everything. Windows users need symlinks enabled or the mirror folders check out as one-line pointer files.
Good for
- Picking individual domain skills (compliance, product, marketing, engineering) from one maintained source
- Teams using several different coding agents who want one library mirrored across them
- Founders wanting C-level advisor personas and commands
Not for
- People who want one deep, tested workflow rather than hundreds of general ones
- Windows users who cannot enable symlinks
What's inside depth 3/5
About 467 skills averaging 7.8 KB with reference docs and many Python helper scripts per the README, plus 59 eval-related files, but the breadth means most individual skills are moderate in depth.
- Skills
- 467
- Reference docs
- 209
- Scripts
- 155
- Tests / evals
- 61
- Templates
- 134
- Avg SKILL.md
- 7.6 KB
Counted from the repository file tree. How we score depth.
Dependencies and costs
- Python 3 for the helper scripts (stdlib only per the README)
- Windows: git symlinks enabled (Developer Mode) and PYTHONUTF8=1
- Some workflow commands expect external MCP servers or accounts depending on the domain
- Telemetry: none found in the README excerpt we read.
Security review
Review before installingA very large library (hundreds of skills, agents and Python tools) of which we only have the README and a set of Gemini skill pointer files. The README offers several installers, including a process-substitution bash <(curl ...) run from the main branch and an unpinned npx CLI, and none of the scripts or the Python tools could be inspected. Nothing in the files we have declares allowed-tools or hooks.
Worth knowing
- Installs with curl | shThe OpenClaw install is
bash <(curl -s https://raw.githubusercontent.com/alirezarezvani/claude-skills/main/scripts/openclaw-install.sh), which runs a script from the main branch unreviewed.README.md:101 - Runs remote code at an unpinned versionThe Codex install runs
npx agent-skills-cli add alirezarezvani/claude-skills --agent codexwith no version pin.README.md:94 - Relies on a CLI or package we could not reviewInstall relies on the agent-skills-cli npm package and on repository shell scripts (scripts/install.sh, convert.sh, openclaw-install.sh) that are not among our files.README.md:94
- Contains compiled binariesThe repo contains two zip archives (ceo-advisor.zip, cto-advisor.zip) that we cannot read.c-level-advisor/ceo-advisor.zip:1
What we checked
The repository (5805 files) on Oct 29, 2026. Not a full review: 5763 of 5805 files were not downloaded (large repository; we reviewed the skill folders); it contains compiled binaries or archives we cannot read; it relies on code outside the repository: scripts/openclaw-install.sh, scripts/install.sh, scripts/convert.sh, scripts/codex-install.sh and the ~964 code files the README describes as 706-774 stdlib-only Python tools (5,763 of 5,805 files were not downloaded); npm package agent-skills-cli; c-level-advisor/ceo-advisor.zip and cto-advisor.zip.
NVIDIA SkillSpector static scan: 4 pattern hits (Anti-Refusal 1, Privilege Escalation 1, MCP Rug Pull 1, Excessive Agency 1). The four hits are README text: 'Developer Mode enabled' in a Windows symlink note, a .env mention in a skill table, an unpinned npx command and 'skip confirmation' for the installer's --force flag. Only the README and 40 one-line Gemini skill pointers were available.
A review, not a certification. Skills change after we check them — read the files before installing.
License
MIT — commercial use allowed.
Can it run unattended?
Not really. The skills are mostly interactive advisory and task playbooks; the README does not describe loops with stop conditions or resumable state.
Alternatives to Claude Skills (alirezarezvani)
Full-Stack Dev Skills (Jeffallan)
Jeffallan/claude-skills67 role-based developer skills for languages, frameworks, DevOps and security, each with reference files, plus 9 Jira/Confluence workflow commands.
Recommended
Check first Depth 3/5
wshobson/agents
wshobson/agentsA multi-harness plugin marketplace with 94 plugins, 202 agents, 185 skills and 105 commands for Claude Code, Codex, Cursor, OpenCode, Copilot, Antigravity and Pi.
With caveats
Check first Depth 2/5
RampStack Claude Skills
rampstackco/claude-skills107 stack-agnostic skills covering the website lifecycle: brand, design, content, SEO, product, ops and growth, with 516 reference files.
With caveats
Check first Depth 3/5
FAQ
What is alirezarezvani/claude-skills?
It is an open-source library of about 400 Claude Code skills, agents, personas and commands covering engineering, marketing, product, compliance and C-level advisory. The same content is mirrored for Codex, Gemini CLI, Cursor and other agents.
How do I install alirezarezvani/claude-skills?
In Claude Code, add the marketplace with /plugin marketplace add alirezarezvani/claude-skills and install a domain bundle such as engineering-skills@claude-code-skills. For Gemini CLI there is a setup script, and for Hermes Agent and Mistral Vibe you run a local sync script.
Is alirezarezvani/claude-skills safe?
The static scan found only documentation-level matches in the README, and the license is MIT. Its Python tools were not part of what was scanned, so review a script before you run it.
Should I install every skill in this repo?
No. Loading hundreds of skills adds descriptions to the agent's context and many overlap. Install the domain bundle you need, or copy individual skills.
Sources: GitHub repository (stars, license, file tree, last push Oct 29, 2026). Review based on the README and SKILL.md files as published.