← Tool directory
TOOL PROFILE

Screaming Frog Headless MCP

Give a local assistant access to headless Screaming Frog crawls and filtered exports from its saved crawl database.

Open source

Community support

Upvotes come from SearchEngineAnswer members. GitHub stars and reader reviews are counted separately.

Claim this tool if you maintain it.

Best for
MCP servers
Setup required
Use Python 3.10 or later, uv, a local MCP client and a licensed Screaming Frog installation in database storage mode.
Costs to check
The community wrapper is MIT-licensed. Screaming Frog requires a paid license for most headless features used by this workflow.
License in catalog
MIT

GitHub repository

GitHub stars
96
Forks
19
Last push
2026-11-14
Primary language
Python

GitHub snapshot from . Stars count GitHub bookmarks. Reader ratings appear separately below.

How to install Screaming Frog Headless MCP

  1. Install and license Screaming Frog SEO Spider, select Database Storage mode, and close the GUI before the wrapper accesses saved crawl data.
    uv tool install screaming-frog-mcp
  2. Run uv tool install screaming-frog-mcp. Register the installed executable as a local stdio server in your MCP client.
  3. Set SF_CLI_PATH if the default path does not match your system. Limit allowed domains with SF_ALLOWED_DOMAINS for automated crawls.

First tutorial for Screaming Frog Headless MCP

  1. Save a small crawl with a known broken URL and preserve a response-code export as your control. Close the GUI.
  2. Run sf_check, list the saved crawls, and ask for response-code rows matching 404 in the selected crawl.
  3. Compare both the count and URLs with the control export. If they differ, confirm you selected the same crawl and report before drawing a conclusion.

Proposed test based on official repository documentation. This community wrapper is separate from the vendor’s built-in MCP integration.

Official installation and usage documentation ↗

Where Screaming Frog Headless MCP fits

Use this community MCP wrapper to inspect saved Screaming Frog crawl data through a local agent or a scheduled audit workflow.

When to choose another tool
Choose the official Screaming Frog MCP when you need its interactive features and vendor-maintained setup instead of the community wrapper.
First evaluation
Open an existing small crawl and count 404 responses in its response-code export. Close the GUI, then ask the MCP server for the same count.
How to interpret the result
Compare the crawl identity, saved filters and exact URL list. A database lock or a missing crawl is an environment issue to resolve before using the result.

Reviews and discussion

Reader reviews

Be the first to share your experience with this tool.

Discussion

Ask a question or share a setup tip.

Write a review or join the discussion

Reviews should describe your own experience. All submissions are moderated.

Flag an outdated detail

Include the field that needs updating and a source link. Copy your draft, then use the corrections policy to contact the editorial team.