Integrations · MCP client
Add SEOBolt to Claude Desktop's MCP config and ask for audits, issue lists, fix plans, keyword difficulty, and Search Console insights in a normal chat.
Available on every plan, including Free
In SEOBolt, open Dashboard → Settings → API Keys and create a key (sk_live_…).
Open ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows) and add the stdio server:
{
"mcpServers": {
"seobolt": {
"command": "npx",
"args": ["-y", "mcp-server-seobolt"],
"env": { "SEOBOLT_API_KEY": "sk_live_YOUR_KEY" }
}
}
}Claude Desktop's config file launches local commands, so bridge to the remote server with mcp-remote. The header value sits in env because a space inside args breaks on Windows. Replace sk_live_YOUR_KEY with a key from Dashboard → Settings → API Keys.
{
"mcpServers": {
"seobolt": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://seobolt.app/mcp", "--header", "Authorization:${AUTH_HEADER}"],
"env": { "AUTH_HEADER": "Bearer sk_live_YOUR_KEY" }
}
}
}Quit and reopen the app; the SEOBolt tools appear in the tools menu.
| MCP tool | What it does |
|---|---|
| seo_projects | List all SEO projects with health scores and stats |
| seo_set_competitors | Set a project's competitor list (domains or names, max 20). |
| seo_audit | Get the SEO audit (health score, page count, issues summary) for a project |
| seo_issues | Get detailed SEO issues with fix recommendations, grouped by type, with affected URLs |
| seo_start_crawl | Start a new SEO crawl for a project. |
| seo_fix_plan | Get the full prioritized fix plan for a project: every site-wide issue with a ready-to-apply code snippet and the affected URLs. |
| seo_fix_suggestions | Get specific code fix suggestions with HTML snippets for a particular page URL |
| seo_content_score | Score a page's SEO optimization for a target keyword. |
| seo_compare | Compare two pages head-to-head on SEO metrics. |
| seo_get_checkout_url | Get a checkout URL to upgrade to a paid plan. |
| seo_domain_rating | Get the Domain Rating (DR) for any domain. |
| seo_keyword_difficulty | Check how hard it is to rank for a keyword. |
| seo_content_gap | Find topics that competitors rank for but you don't. |
| seo_link_prospects | Find domains that link to competitors but not to you. |
| seo_gsc_content_opportunities | Find keywords a site ALREADY ranks for in Google Search Console but does not target on-page — the highest-ROI content play. |
| seo_gsc_reality | Explain WHY a site's top pages do (or don't) get traffic — fuses each page's real GSC numbers (impressions/clicks/position) with a live on-page crawl and returns an honest per-page verdict plus a single headline naming the true growth bottleneck (on-page hygiene vs. authority/rank vs. snippet/CTR). |
| seo_gsc_prune_candidates | Find pages worth pruning from a connected Google Search Console property: 0 clicks and <50 impressions over the window (default 90 days). |
| seo_gsc_repoint | Find pages whose top Google Search Console query is not reflected in the live title/H1 (connected GSC + live crawl). |
| seo_gsc_cannibalization | Find keyword cannibalization in a connected Google Search Console property: queries where 2+ URLs of the site each get at least 10% of the query's impressions. |
| seo_gsc_secondary_keywords | List every query a page already ranks top-10 for in a connected Google Search Console property, grouped by overlap, with a suggested title (≤60 chars) covering the top 2-3 secondary queries. |
| seo_gsc_content_decay | Find pages losing traffic in connected Google Search Console: the last 28 days (or `days`) vs the window before, per page. |
| seo_gsc_low_ctr | Find pages and queries in connected Google Search Console with high impressions but a CTR well below what their average position should earn (expected-CTR curve by position from public CTR studies): ≥200 impressions, position ≤20, CTR under half the expected value. |
| seo_gsc_index_coverage | Check whether Google has indexed your pages, using the Search Console URL Inspection API (real Google data, free quota 2,000/day per property). |
| seo_core_web_vitals | Core Web Vitals (mobile) from Google PageSpeed Insights for a URL, or for the first pages (≤5) of a project's latest crawl. |
| seo_brand_mentions | Brand Radar: find pages that mention your brand or domain without linking to you (unlinked mentions = easy link requests). |
| seo_find_contact | Find how to contact a website for outreach: emails published on the page and on the same site's /contact, /contact-us, /about and /write-for-us pages (mailto links, visible text, Cloudflare-protected addresses), plus its contact page / contact form URL. |
| seo_ai_citations | AI citation check (GEO): ask an AI engine with live web search (OpenAI Responses API + web_search) up to 10 buyer prompts and report, per prompt, whether your domain is cited, whether your brand is mentioned, which other domains are cited, and an answer excerpt. |
No. API keys, the REST API, and the MCP server work on every plan, including Free. What each tool can do follows your plan's limits: the number of crawls, pages per crawl and project count depend on the plan, Brand Radar tools (brand mentions and contact lookups) need Starter or above, and AI citation checks and Google Search Console tools need Pro or above. The API key comes from Dashboard → Settings → API Keys.
Both expose the same tools. The remote endpoint (https://seobolt.app/mcp) needs nothing installed. The npx package (mcp-server-seobolt) runs locally over stdio and needs Node.js.
seo_fix_plan returns each issue with a code snippet and the affected URLs. A coding agent can apply those to your repository; you deploy, then seo_start_crawl and seo_audit confirm the score went up.
The key is sent only to seobolt.app as a Bearer token. Keep the config file out of version control and revoke the key in Settings if it leaks.
Not yet. The remote endpoint takes a Bearer API key and does not offer OAuth sign-in, which one-click connectors in claude.ai, Claude Desktop and ChatGPT require. Use the config above.