Turbot Guardrails Model Context Protocol (MCP) Server

@turbot

About Turbot Guardrails Model Context Protocol (MCP) Server

Enable AI assistants to explore and query your Turbot Guardrails data!

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "guardrails-mcp": {
      "command": "npx",
      "args": [
        "@modelcontextprotocol/inspector",
        "node",
        "dist/index.js"
      ]
    }
  }
}

Tools

No tools detected

Fetch the live tool list by running this server in a temporary sandbox using the button above.

Overview

What is Turbot Guardrails Model Context Protocol (MCP) Server?

This MCP server connects AI assistants such as Claude to a Turbot Guardrails environment, enabling natural language exploration, analysis, and automation across cloud resources. It is designed for cloud engineers and governance teams who want to query, control, and monitor their cloud estate through conversational interfaces.

How to use Turbot Guardrails Model Context Protocol (MCP) Server?

Install via npm (npx -y @turbot/guardrails-mcp) and configure using either a Turbot CLI profile (set TURBOT_PROFILE) or direct environment variables (TURBOT_WORKSPACE, TURBOT_ACCESS_KEY, TURBOT_SECRET_KEY). Legacy v0.1.x variable names are also accepted. Add the configuration to your AI assistant's MCP settings file (e.g., claude_desktop_config.json for Claude Desktop) and restart the assistant.

Key features of Turbot Guardrails Model Context Protocol (MCP) Server

Run read-only GraphQL queries against Guardrails data.
List and filter resources, controls, and policy types.
Execute controls by ID for compliance remediation.
Introspect GraphQL schemas of runnable types.
Render Nunjucks templates for dynamic configurations.

Use cases of Turbot Guardrails Model Context Protocol (MCP) Server

Query cloud resources (e.g., S3 buckets, EC2 instances) using natural language.
Generate compliance and security reports on non-compliant resources.
Explore available policy and control types by category or title.
Run controls on specific resources directly from the AI assistant.
Process and apply Nunjucks templates for dynamic policy or configuration generation.

FAQ from Turbot Guardrails Model Context Protocol (MCP) Server

How do I authenticate to Guardrails?

You can authenticate either by referencing a Turbot CLI profile with TURBOT_PROFILE or by setting the three direct credential variables: TURBOT_WORKSPACE, TURBOT_ACCESS_KEY, and TURBOT_SECRET_KEY. Legacy v0.1.x names (e.g., TURBOT_GRAPHQL_ENDPOINT) are still accepted.

What are the prerequisites?

Node.js version 20 or higher, a valid Guardrails API access key with appropriate permissions, and the endpoint URL for your Guardrails workspace.

Where does the server look for credentials by default?

If using a CLI profile, the server reads ~/.config/turbot/credentials.yml by default. This path can be overridden with the TURBOT_CLI_CREDENTIALS_PATH environment variable.

What should I do if I see authentication errors?

Verify that your API key is correct and has the necessary permissions. Ensure the profile name matches an entry in your credentials file and that the file includes workspace, accessKey, and secretKey. Credential values are redacted from error messages returned to your AI assistant.

How can I test the server locally?

Use the MCP Inspector by running npm run build then npx @modelcontextprotocol/inspector node dist/index.js. This launches an interactive test interface for the server.

Frequently asked questions

How do I authenticate to Guardrails?

You can authenticate either by referencing a Turbot CLI profile with `TURBOT_PROFILE` or by setting the three direct credential variables: `TURBOT_WORKSPACE`, `TURBOT_ACCESS_KEY`, and `TURBOT_SECRET_KEY`. Legacy v0.1.x names (e.g., `TURBOT_GRAPHQL_ENDPOINT`) are still accepted.

What are the prerequisites?

Node.js version 20 or higher, a valid Guardrails API access key with appropriate permissions, and the endpoint URL for your Guardrails workspace.

Where does the server look for credentials by default?

If using a CLI profile, the server reads `~/.config/turbot/credentials.yml` by default. This path can be overridden with the `TURBOT_CLI_CREDENTIALS_PATH` environment variable.

What should I do if I see authentication errors?

Verify that your API key is correct and has the necessary permissions. Ensure the profile name matches an entry in your credentials file and that the file includes `workspace`, `accessKey`, and `secretKey`. Credential values are redacted from error messages returned to your AI assistant.

How can I test the server locally?

Use the MCP Inspector by running `npm run build` then `npx @modelcontextprotocol/inspector node dist/index.js`. This launches an interactive test interface for the server.

Comments

Basic information

More Other MCP servers

Codelf

unbug

A search tool helps dev to solve the naming things problem.

14.2K

MCP Registry

modelcontextprotocol

A community driven registry service for Model Context Protocol (MCP) servers.

7.0K

🚀 Model Context Protocol (MCP) Curriculum for Beginners

microsoft

This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,

16.7K

Activepieces

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

23.2K

Glasswarp

See and control a real Windows PC you own — from any MCP client, locally or remotely. Observe (UIA + screenshots), click/type/drag/scroll, launch apps, owner Live View. BYOH: your machine, your key.