What it does
geoscore checks how well a website is set up for AI crawlers and search, a practice the publisher calls Generative Engine Optimization (GEO). The analyze_geo_score tool grades an existing domain from 0 to 100 and breaks the result into crawlability, robots.txt rules for eight named AI crawlers, visibility, Schema.org, FAQ, contacts and hygiene (freshness, accessibility and llms.txt). It is built around five tools.
Someone would reach for it to fix an existing site or to set up a new one. get_geo_fixes returns a markdown document with a robots.txt block, an Organization JSON-LD template, a FAQPage skeleton and an llms.txt skeleton for a domain. make_site_geo_ready produces the same kinds of assets for a new site from a name, URL and description, with optional services, email and FAQ entries. suggest_monitoring_config drafts a monitoring setup with target audiences and suggested queries, and get_site_report returns a summary, using a causabi.com share slug for a full mention-rate breakdown or a domain alone for the crawlability score.
It runs locally over stdio and needs no account and no API key. Site facts are fetched directly from the domain being checked, with a fixed 15-second timeout per request. It can be used through an MCP client configuration, a command-line interface that takes a JSON argument, or Docker, and it is released under the MIT licence.
Limitations and requirements
The publisher says the original server's sixth tool, get_product_guide, is omitted because its content is not published. The scoring weights are a reconstruction documented by the publisher, not the original's. The causabi.com report API path is also a reconstruction. suggest_monitoring_config makes no API call; applying its config needs the PUT calls on causabi.com with a site_id placeholder filled in. A full report requires a causabi.com report slug; a domain alone gives only the crawlability score. If a home page cannot be fetched, analyze_geo_score fails with a site unreachable error.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.CloudCorpRecords/geoscore
AI-crawler readiness checks and fixes: GEO score, robots.txt, schema.org, FAQ and llms.txt.
v0.1.0As shown on registry.modelcontextprotocol.io when we last read it