# AgentTwin Documentation

> AgentTwin Documentation is a free MCP server for developer tools. Search AgentTwin's agent-testing guides, CLI commands and service contracts over MCP. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Free
- **Platforms:** API, Web
- **Website:** https://agenttwin.si/developers/
- **MCP Registry name:** io.github.AgentTwin/agenttwin-docs
- **Categories:** developer tools, AI models and LLM tooling
- **Listing updated:** 2026-11-15

## About

This is a public, read-only documentation server for AgentTwin, a product that builds digital twins for testing AI agents against synthetic data. It needs no credentials and offers two tools, search_docs and read_doc, plus documentation resources. It cannot reach accounts or private repositories, and it cannot run workflows.

The endpoint speaks MCP protocol versions 2025-03-26, 2025-06-18 and 2025-11-25 with JSON responses, without standalone SSE, and is limited to 600 requests a minute per instance. The same corpus is searchable through an NLWeb list-mode endpoint, and WebMCP-capable browsers get matching page tools.

The docs cover the Node.js CLI (npx agenttwin), a local stdio MCP mode, the REST API with scoped OAuth client credentials, and supported synthetic service contracts. AgentTwin itself is sold as an enterprise offering; its pricing page asks buyers to make contact.

## Features

- **Documentation search:** search_docs finds guide sections on agent testing and CLI usage.
- **Document reading:** read_doc returns a full published page.
- **No credentials:** The endpoint is open and read-only.
- **Service contract coverage:** Docs describe supported provider operations, safety and status.
- **NLWeb endpoint:** The same docs are queryable by POST to /ask.
- **Rate-limit headers:** Responses carry RateLimit headers and Retry-After on 429.

## Good for

- Look up CLI commands while setting up agent tests
- Check which synthetic service operations AgentTwin supports
- Find authentication steps for the REST API
- Let a coding agent read AgentTwin guides before preparing a workflow
- Compare the local stdio MCP with the public docs MCP

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http agenttwin-docs https://agenttwin.si/mcp/docs
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://agenttwin.si/mcp/docs
```

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

```json
{
  "mcpServers": {
    "agenttwin-docs": {
      "url": "https://agenttwin.si/mcp/docs"
    }
  }
}
```

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

```json
{
  "servers": {
    "agenttwin-docs": {
      "type": "http",
      "url": "https://agenttwin.si/mcp/docs"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "agenttwin-docs": {
      "serverUrl": "https://agenttwin.si/mcp/docs"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "agenttwin-docs": {
      "type": "http",
      "url": "https://agenttwin.si/mcp/docs"
    }
  }
}
```

## Tools

- `search_docs`: Searches AgentTwin's published documentation.
- `read_doc`: Reads a published documentation page.

## FAQ

### Do I need an account?

No. The documentation endpoint is public and needs no credentials.

### Can it run tests?

No. It cannot access accounts or private repositories or run workflows.

### What is the endpoint?

https://agenttwin.si/mcp/docs over streamable HTTP.

### What is the rate limit?

600 requests per minute, aggregated per instance.

### How do I run the real testing tools?

Use the Node CLI with npx agenttwin, which also has a local stdio MCP mode.

### Is AgentTwin Documentation free?

Yes. AgentTwin Documentation is free to use.

### How do I add AgentTwin Documentation to Claude Code?

Run this command in your terminal: claude mcp add --transport http agenttwin-docs https://agenttwin.si/mcp/docs

### What are the best alternatives to AgentTwin Documentation?

The closest alternatives to AgentTwin Documentation are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.

## Alternatives

- [REA](https://ai-powered.com/mcp/rea): REA is an open-source MCP server for developer tools.
- [Codebase Memory](https://ai-powered.com/mcp/codebase-memory): Codebase Memory is an open-source MCP server for knowledge and memory.
- [GitHub](https://ai-powered.com/mcp/github-2): GitHub is an open-source MCP server for developer tools.
- [Serena MCP: the IDE for your agent](https://ai-powered.com/mcp/serena-mcp-the-ide-for-your-agent): Serena MCP: the IDE for your agent is an open-source MCP server for developer tools.
- [XERJ](https://ai-powered.com/mcp/xerj): XERJ is an open-source MCP server for knowledge and memory.
- [Ouroboros](https://ai-powered.com/mcp/ouroboros): Ouroboros is an open-source MCP server for developer tools.
- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [Concord MCP](https://ai-powered.com/mcp/concord-mcp): Concord MCP is a freemium MCP server for developer tools.

---
Source: https://ai-powered.com/mcp/agenttwin-documentation
