Use when the user wants to find, evaluate, audit, or install an open-source AI agent skill or MCP server — e.g. "find an MCP server for Postgres", "is this skill safe to install", "what should I use to scrape a website". Searches a quality-scored, security-graded catalog of ~20K skills locally (cached index, zero backend load) and returns each result's security grade, quality score, and install commands so you can check trust BEFORE installing.
npx mdskills install zhuyansen/agentskillshub-cliWell-structured skill with clear trigger conditions and security-first workflow guidance
🌐 Website · 🔎 Browse Skills · 🛡️ Security Report · 🏢 Enterprise
Search, audit, and install open-source AI agent skills & MCP servers from your terminal. Every result is security-graded and quality-scored by AgentSkillsHub — a directory of 117K+ skills with a trust signal before you install.
npx @agentskillshub/cli search "scrape a website" --safe
Discovering a skill is easy. Knowing whether it's safe to run against your credentials is not. It puts the trust signal before the install:
$ npx @agentskillshub/cli search postgres --category mcp-server --limit 2
call518/MCP-PostgreSQL-Ops 150★ 🟢 SAFE
Give AI assistants full PostgreSQL DBA superpowers — 30+ tools…
mcp-server · claude-code · mcp · python https://agentskillshub.top/skill/call518/MCP-PostgreSQL-Ops/
sgaunet/postgresql-mcp 5★ ⚪ UNAUDITED ~23.0k tok
A Model Context Protocol (MCP) server… read-only query execution…
mcp-server · claude-code · go · mcp https://agentskillshub.top/skill/sgaunet/postgresql-mcp/
Run via npx @agentskillshub/cli :
search [filters] | Find skills. Filters: --category --platform --min-stars --safe --limit |
audit | Free basic trust check: security grade + plain-English verdict |
install | Install commands + "check before you install" safety line |
update | Force-refresh the cached index |
Add --json to any of search / audit / install for machine-readable output.
The catalog (~20K quality skills, stars ≥ 5) is a single static file (~1.7MB gzipped) served from the CDN. The CLI downloads it once, caches it at ~/.cache/agentskillshub/, and checks for a newer index with a cheap 77B probe (re-downloading the full file only when it actually changed, so a freshly-deployed index reaches you within minutes). All searching is local — fast, works offline, and puts zero load on the backend.
🟢 SAFE · 🟡 CAUTION · 🔴 UNSAFE · ⛔ REJECT · ⚪ UNAUDITED
⚪ UNAUDITED is not "probably fine" — it means no one has audited it. Check the code, the credentials it asks for, and who maintains it before you trust it.
We scanned the whole catalog and wrote up what we found — only 17.7% of skills are popular enough to be graded, 1 in 32 graded ones is unsafe, and the risk lives in the long tail: We security-graded 117,854 AI agent skills.
| Var | Default |
|---|---|
AGENTSKILLSHUB_BASE | https://agentskillshub.top |
AGENTSKILLSHUB_CACHE | ~/.cache/agentskillshub |
NO_COLOR | unset (set to disable color) |
MIT © AgentSkillsHub
Install via CLI
npx mdskills install zhuyansen/agentskillshub-cliAgent Skills Hub is a free, open-source AI agent skill. Use when the user wants to find, evaluate, audit, or install an open-source AI agent skill or MCP server — e.g. "find an MCP server for Postgres", "is this skill safe to install", "what should I use to scrape a website". Searches a quality-scored, security-graded catalog of ~20K skills locally (cached index, zero backend load) and returns each result's security grade, quality score, and install commands so you can check trust BEFORE installing.
Install Agent Skills Hub with a single command:
npx mdskills install zhuyansen/agentskillshub-cliThis downloads the skill files into your project and your AI agent picks them up automatically.
Agent Skills Hub works with Claude Code, Claude Desktop, Cursor, Vscode Copilot, Windsurf, Continue Dev, Codex, Gemini Cli, Amp, Roo Code, Goose, Opencode, Trae, Qodo, Command Code. Skills use the open SKILL.md format which is compatible with any AI coding agent that reads markdown instructions.