crawlcove-mcp
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| crawl_siteA | Crawl a website live (same-origin, breadth-first, robots.txt respected) and make the result the active dataset for the other tools. Capped at 200 pages; for a full-site audit with history and Search Console data, use the Crawl Cove desktop app and load its export with load_export. |
| load_exportA | Load a Crawl Cove desktop app JSON export (Reports → Export → JSON) or a crawlcove-cli JSON result from disk and make it the active dataset. Desktop exports carry more per-page data (depth, word count, content type) and have no page cap. |
| get_issuesA | List the SEO issues found in the active dataset, optionally one type only. Types: broken-page (The page returned a 4xx/5xx status or could not be fetched); missing-title (No ); long-title ( longer than 60 characters (may be truncated in search results)); duplicate-title (The same is used on more than one page); missing-meta-description (No ); long-meta-description (Meta description longer than 160 characters); missing-h1 (No on a 200 HTML page); multiple-h1 (More than one ); noindex (The page carries a robots noindex directive); redirect-chain (The URL went through 2 or more redirects); missing-canonical (No on an indexable page). |
| get_pageA | Return everything the crawl recorded about one URL (status, title, meta description, H1 count, canonical, robots, redirects, broken links out). |
| list_broken_linksA | Broken internal links as source page → broken target pairs, so you know which page to fix. Needs a live crawl or crawlcove-cli dataset; a desktop export lists the broken pages instead (its link graph lives in the app). |
| list_pagesA | List crawled pages with status and title, optionally filtered by status code, indexability, or a URL substring. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 6 tools
crawl_site and load_export both establish the active dataset but are clearly differentiated by source (live crawl vs. file import). The four query tools target distinct resources (issues, page details, broken link pairs, page list) with no meaningful overlap.
All tools use snake_case verb_noun, with a small set of verbs (crawl, load, get, list) that accurately reflect actions. No inconsistent naming conventions.
Six tools is well-scoped for a crawl-audit MCP: two ingestion paths, two listing/query tools, and two issue-focused tools. No obvious bloating or missing essential utility tools.
Core lifecycle is covered: ingest (live or export), list pages, inspect a page, list issues, and trace broken links. A site-level summary/statistics tool or a way to clear/switch datasets would round it out, but agents can work around these gaps.