Skip to content
MCP server

Automation Auditor MCP

Static checks for n8n, Dify, Trigger.dev and Activepieces exports before deploying

Runs locallyStatus and history
Install
Pricing
Open source
Platforms
CLI, self-hosted
Runs
Locally
Version
0.1.0
Listing updated
2 days ago

About Automation Auditor MCP

Automation Auditor MCP is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.

This server never touches your repository or your automation accounts. You paste in the text of an export, and it runs read-only static checks on that text: no JavaScript is executed, no API endpoints are called, and no files are read by path. Input is capped at 250,000 characters.

Coverage depends on the platform. For n8n it looks for missing or duplicate node names and IDs, broken connections and missing webhook paths. Dify exports are checked for node ID problems, broken edges and start/end nodes. Trigger.dev source gets heuristic checks on literal task IDs and duplicates, and Activepieces flows are checked for a trigger, duplicate step names and steps marked invalid. Platforms without a native adapter are reported as unsupported instead of getting a meaningless pass, and Make, Zapier and Pipedream are not supported in v0.1.

Findings carry a stable rule ID, severity, location, short message, recommendation and confidence. They deliberately leave out node names, raw URLs, credentials and source snippets. The README warns that a result of no findings is not proof of safety and that this is an early-stage release to verify against your MCP host before production use.

Install

Run in a terminal

claude mcp add automation-auditor-mcp -- uvx automation-auditor-mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "automation-auditor-mcp": {
      "command": "uvx",
      "args": [
        "automation-auditor-mcp"
      ]
    }
  }
}
Install in CursorOpens Cursor, which asks you to confirm. Or add it by hand:

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "automation-auditor-mcp": {
      "command": "uvx",
      "args": [
        "automation-auditor-mcp"
      ]
    }
  }
}
Install in VS CodeOpens VS Code, which asks you to confirm. Or add it by hand:

Add to .vscode/mcp.json

{
  "servers": {
    "automation-auditor-mcp": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "automation-auditor-mcp"
      ]
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "automation-auditor-mcp": {
      "command": "uvx",
      "args": [
        "automation-auditor-mcp"
      ]
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "automation-auditor-mcp": {
      "command": "uvx",
      "args": [
        "automation-auditor-mcp"
      ]
    }
  }
}

Tools this server provides (4)

Taken from the project documentation; the running server may differ.

audit_workflow
Audits the text of an n8n, Dify, Trigger.dev or Activepieces export for the given platform.
audit_openapi
Checks a JSON OpenAPI document.
preflight_url
Screens a URL for unsafe schemes and addresses without DNS or HTTP requests.
list_supported_platforms
Lists supported formats and known limitations.

What it does

Multi-platform workflow audit
Checks n8n, Dify, Trigger.dev and Activepieces exports through one audit_workflow tool.
Structural checks
Flags missing or duplicate node IDs, broken connections or edges, and missing triggers.
OpenAPI checking
Runs a best-effort integration check on a JSON OpenAPI spec.
URL preflight
Screens URLs for unsafe schemes and addresses without DNS lookups or fetching.
Redacted findings
Reports rule ID, severity and location without echoing credentials, URLs or snippets.
Honest platform list
list_supported_platforms returns the supported formats and their known limits.

Good for

  • Catch broken connections in an n8n export before importing it
  • Review a Dify workflow DSL for missing start or end nodes
  • Scan Trigger.dev task definitions for duplicate task IDs
  • Check an Activepieces flow for a missing trigger
  • Screen a URL for unsafe schemes before an agent uses it

Pricing

Open source

Prices change. Check ahmadyar.co before you buy.

Automation Auditor MCP FAQ

Does it read my files or call my automation accounts?

No. It only analyses the text you pass in and never reads files by path, contacts accounts or connects to API endpoints.

Which platforms are supported?

n8n, Dify, Trigger.dev and Activepieces, plus OpenAPI JSON and URL preflight. Make, Zapier and Pipedream are not supported in v0.1.

How do I install it?

Run pip install automation-auditor-mcp with Python 3.11+, or point an MCP client at uvx automation-auditor-mcp.

Is a clean report a guarantee?

No. The README states that no findings detected is not proof of safety, and Trigger.dev checks are heuristics only.

Is Automation Auditor MCP free?

Yes. Automation Auditor MCP is open source, so you can use and self-host it for free.

How do I add Automation Auditor MCP to Claude Code?

Run this command in your terminal: claude mcp add automation-auditor-mcp -- uvx automation-auditor-mcp

What are the best alternatives to Automation Auditor MCP?

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

Alternatives to Automation Auditor MCP

  1. REA

    MCP server

    CLI and MCP server that lets coding agents reverse engineer apps and binaries

    Runs locally
    • Open source
    • 464K GitHub stars
    • Developer tools
  2. Codebase Memory

    MCP server

    Local MCP server that builds a code knowledge graph for AI coding agents

    Runs locally
    • Open source
    • 48.7K GitHub stars
    • Knowledge and memory
  3. GitHub

    MCP server

    Official MCP server connecting AI assistants to GitHub repos, issues and workflows

    Not checked yet
    • Remote
    • Open source
    • 34.5K GitHub stars
    • Developer tools
  4. MCP toolkit giving coding agents symbol-level code retrieval and editing

    Runs locally
    • Open source
    • 31.5K GitHub stars
    • Developer tools
  5. XERJ

    MCP server

    Local search engine for AI agents with cited answers over your code and docs

    Runs locally
    • Open source
    • 19.2K GitHub stars
    • Knowledge and memory
  6. Ouroboros

    MCP server

    Spec-first layer that pins requirements and verifies results of AI coding agent work

    Runs locally
    • Open source
    • 6.4K GitHub stars
    • Developer tools