# TrustDex MCP server

> Provides a local-first trust and provenance gate to evaluate and filter MCP servers, skills, and plugins for AI agents.

- Repository: https://github.com/grigent/trustdex
- Page: https://hostedmcp.org/servers/grigent/trustdex
- Transport: stdio
- Stars: 1
- License: MIT

## Install

### Claude Code

```sh
claude mcp add trustdex -- npx -y trustdex
```

### Cursor (`~/.cursor/mcp.json`)

```json
{
  "mcpServers": {
    "trustdex": {
      "command": "npx",
      "args": [
        "-y",
        "trustdex"
      ]
    }
  }
}
```

### VS Code (`.vscode/mcp.json`)

```json
{
  "servers": {
    "trustdex": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "trustdex"
      ]
    }
  }
}
```

### Claude Desktop (`claude_desktop_config.json`)

```json
{
  "mcpServers": {
    "trustdex": {
      "command": "npx",
      "args": [
        "-y",
        "trustdex"
      ]
    }
  }
}
```

### Codex (`~/.codex/config.toml`)

```toml
[mcp_servers.trustdex]
command = "npx"
args = ["-y", "trustdex"]
```

### Gemini CLI (`~/.gemini/settings.json`)

```json
{
  "mcpServers": {
    "trustdex": {
      "command": "npx",
      "args": [
        "-y",
        "trustdex"
      ]
    }
  }
}
```

## FAQ

### What is TrustDex's primary function?

TrustDex acts as a decision layer that evaluates MCP servers, skills, and plugins to determine if they should be allowed, require human review, or be blocked.

### Is TrustDex a malware scanner?

No, it is security-related infrastructure designed for trust and provenance, not a malware scanner, sandbox, or certification service.

### How does TrustDex handle tool exposure to agents?

It can generate a filtered configuration file that only contains tools that have been explicitly approved by your policy.

### What are the built-in policy options?

TrustDex includes three built-in policy packs: strict, official-first, and development.