MCP Server

AgenticStack for MCP

Connect any MCP-compatible agent to the AgenticStack tool directory. Search, compare, and get recommendations without leaving your editor or chat interface. No API key required.

Install

Requires Node 18+. Add the server to your MCP client config:

Claude Code

Run once in your terminal

claude mcp add agenticstack -- npx -y agenticstack-mcp

Claude Desktop

~/Library/Application Support/Claude/claude_desktop_config.json

{
  "mcpServers": {
    "agenticstack": {
      "command": "npx",
      "args": ["-y", "agenticstack-mcp"]
    }
  }
}

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "agenticstack": {
      "command": "npx",
      "args": ["-y", "agenticstack-mcp"]
    }
  }
}

Tools

The server exposes five tools. A typical flow: call list_categories first to discover valid slugs, then search_tools or recommend_tool to find candidates, then get_tool or compare_tools to go deeper.

list_categories

Lists all tool categories with descriptions, feature definitions, and tool slugs. Start here to discover valid category slugs and agent_features keys.

search_tools

Filters the directory by category, language, framework, or feature. Returns a summary list of matching tools.

get_tool

Full details for a single tool — agent features, pricing tiers, SDK languages, compliance certifications, limitations, and editorial analysis.

compare_tools

Side-by-side comparison of 2–3 tools, including a feature matrix and editorial analysis where available.

recommend_tool

Scores and ranks tools by how well they fit your requirements. Returns the top 3 with a match_score. Use this when you want a recommendation, not just a filtered list.

Example prompts

  • What auth tools support MCP and work with Python?
  • Compare Auth0, Clerk, and WorkOS for an agentic app
  • Recommend an observability tool for a TypeScript LangChain project
  • What categories does AgenticStack cover?