# Agent Connectors MCP server

Read-only MCP server that searches the public [Agent Connectors](https://agentconnectors.dev) directory of MCP servers and AI agent connectors.

[![JaniJegoroff/agent-connectors-mcp MCP server](https://glama.ai/mcp/servers/JaniJegoroff/agent-connectors-mcp/badges/score.svg)](https://glama.ai/mcp/servers/JaniJegoroff/agent-connectors-mcp)

[Browse the full directory](https://agentconnectors.dev)

## What it does

The tools read `https://agentconnectors.dev/api/tools.json?limit=all` and follow `next` if the response is paged. They return directory records. They do not run, install, or test the servers in those records.

- `search_tools` — match a query, with an optional category slug, against listing name, slug, and description.
- `get_tool` — one listing by slug.
- `list_categories` — category name, slug, and listing count, counted from the catalogue response.

Example query for `search_tools`:

```json
{ "query": "web search", "category": "search" }
```

## Install

This repository is not published to npm. Clone it, install dependencies, and point your client at the clone. Requires Node.js 22 or newer.

```bash
git clone https://github.com/JaniJegoroff/agent-connectors-mcp.git
cd agent-connectors-mcp
npm install
```

The server speaks MCP over stdin and stdout:

```bash
node --experimental-strip-types src/index.ts
```

Replace `/absolute/path/to/agent-connectors-mcp` with the absolute path of your clone. Run `npm install` in that directory first.

### Claude Desktop

Add this to `claude_desktop_config.json`:

```json
{
  "mcpServers": {
    "agent-connectors": {
      "command": "node",
      "args": ["--experimental-strip-types", "src/index.ts"],
      "cwd": "/absolute/path/to/agent-connectors-mcp"
    }
  }
}
```

### Cursor

Project file `.cursor/mcp.json`, or the global Cursor MCP settings:

```json
{
  "mcpServers": {
    "agent-connectors": {
      "command": "node",
      "args": ["--experimental-strip-types", "src/index.ts"],
      "cwd": "/absolute/path/to/agent-connectors-mcp"
    }
  }
}
```

### VS Code

Project file `.vscode/mcp.json`. `cwd` is the clone root:

```json
{
  "servers": {
    "agent-connectors": {
      "type": "stdio",
      "command": "node",
      "args": ["--experimental-strip-types", "src/index.ts"],
      "cwd": "${workspaceFolder}"
    }
  }
}
```

Optional streamable HTTP, bound to `127.0.0.1:8791`:

```bash
node --experimental-strip-types src/index.ts --http
```

The HTTP endpoint is `http://127.0.0.1:8791/mcp`. Set `PORT` to use another port. Set `AGENT_CONNECTORS_API` to point the catalogue fetch at another `tools.json` URL.

## Docker

Glama builds this image and introspects the server over stdio.

```bash
docker build -t agent-connectors-mcp .
docker run -i --rm agent-connectors-mcp
```

The image is `node:22-slim`. It installs production dependencies with `npm ci --omit=dev`, copies `src`, and starts stdio mode with `node --experimental-strip-types src/index.ts`.