@llmrank.app/mcp

MCP server for LLM Boost — AI-readiness SEO tools for your coding agent. Crawl sites, score pages across 37 factors, track AI visibility, and get actionable fixes to improve how AI assistants cite your content.

Getting Started

1. Get an API token at llmrank.app/dashboard/settings (API Tokens tab)
2. Configure the MCP server in your IDE (see below)

Setup

Claude Code

claude mcp add llm-boost \
  --env LLM_BOOST_API_TOKEN=llmb_xxx \
  -- npx -y @llmrank.app/mcp

Team setup — share the config with your team via .mcp.json (committed to git):

claude mcp add llm-boost --scope project \
  --env LLM_BOOST_API_TOKEN \
  -- npx -y @llmrank.app/mcp

Each team member sets LLM_BOOST_API_TOKEN in their shell environment (e.g. .env, .zshrc). The .mcp.json config is shared via git — no tokens in source control.

Cursor

Add to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "llm-boost": {
      "command": "npx",
      "args": ["-y", "@llmrank.app/mcp"],
      "env": {
        "LLM_BOOST_API_TOKEN": "llmb_xxx"
      }
    }
  }
}

Claude Desktop

Add to claude_desktop_config.json (find config file):

{
  "mcpServers": {
    "llm-boost": {
      "command": "npx",
      "args": ["-y", "@llmrank.app/mcp"],
      "env": {
        "LLM_BOOST_API_TOKEN": "llmb_xxx"
      }
    }
  }
}

VS Code

Add to .vscode/mcp.json in your project:

{
  "servers": {
    "llm-boost": {
      "command": "npx",
      "args": ["-y", "@llmrank.app/mcp"],
      "env": {
        "LLM_BOOST_API_TOKEN": "llmb_xxx"
      }
    }
  }
}

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "llm-boost": {
      "command": "npx",
      "args": ["-y", "@llmrank.app/mcp"],
      "env": {
        "LLM_BOOST_API_TOKEN": "llmb_xxx"
      }
    }
  }
}

HTTP Transport (ChatGPT, remote clients)

For clients that support Streamable HTTP (e.g. ChatGPT, OpenAI Agents SDK), use the hosted endpoint:

Endpoint: https://mcp.llmrank.app/v1/mcp
Auth: Bearer <your_llmb_token>

The endpoint also supports OAuth 2.1 with PKCE — see /.well-known/oauth-authorization-server for discovery.

Perplexity

Perplexity supports local MCP servers on macOS. Add via Settings > Connectors using the same JSON config as Cursor above.

Environment Variables

| Variable | Required | Description | | --------------------- | -------- | ---------------------------------------------------------------------------------------------------------------------- | | LLM_BOOST_API_TOKEN | Yes | API token starting with llmb_ — generate at llmrank.app/dashboard/settings | | LLM_BOOST_API_URL | No | API base URL (default: https://api.llmrank.app) |

Available Tools

Projects

| Tool | Description | | ---------------- | ------------------------------------------------ | | list_projects | List all projects with domains and latest scores | | get_project | Get project details including latest crawl score | | create_project | Create a new project by providing a domain |

Crawling

| Tool | Description | | ------------------ | --------------------------------------------------------------- | | start_crawl | Start a new crawl — scores pages across 37 AI-readiness factors | | get_crawl_status | Check crawl progress (pending → crawling → scoring → complete) | | list_crawls | Get crawl history with scores and timestamps |

Scores & Analysis

| Tool | Description | | ------------------- | ------------------------------------------------------------------------------------------------------------------ | | get_site_score | Overall AI-readiness score with category breakdown (Technical 25%, Content 30%, AI Readiness 30%, Performance 15%) | | compare_scores | Compare two crawls to see improvements or regressions | | get_score_history | Score trends over time across crawls | | list_pages | List crawled pages with scores, sortable and filterable by grade | | get_page_details | Full page analysis: per-category scores, all issues, and fixes |

Issues & Fixes

| Tool | Description | | ------------------------ | ---------------------------------------------------------------------- | | list_issues | All issues grouped by severity (critical/warning/info) and category | | get_fix_recommendation | AI-generated fix steps with code examples and expected score impact | | generate_fix | Generate code snippets and content changes to resolve a specific issue |

AI Visibility

| Tool | Description | | ------------------------- | ------------------------------------------------------------------------------------------------------------------------ | | check_visibility | Check if your brand appears in AI search results across 6 platforms (ChatGPT, Claude, Perplexity, Gemini, Copilot, Grok) | | list_visibility_history | Track AI search presence over time | | suggest_queries | AI-suggested queries to monitor based on your content and industry |

Content & Technical

| Tool | Description | | ------------------- | ----------------------------------------------------------- | | analyze_content | Evaluate page content across 37 AI-readiness factors | | suggest_meta_tags | Generate optimized title, description, and Open Graph tags | | check_llms_txt | Validate your llms.txt file for AI crawler permissions | | validate_schema | Check structured data (JSON-LD, Schema.org) for correctness |

Strategy

| Tool | Description | | --------------------- | ---------------------------------------------------------------- | | get_recommendations | Prioritized action plan ranked by effort and impact | | get_content_gaps | Topics competitors cover that you don't | | discover_keywords | AI-powered keyword discovery with search volume and AI relevance | | list_competitors | Tracked competitors with AI-readiness score comparison | | compare_competitor | Side-by-side comparison with a specific competitor |

Reports

| Tool | Description | | ----------------- | ---------------------------------------------------------------------- | | generate_report | Comprehensive Markdown report with scores, issues, and recommendations |

Prompts

The server includes pre-built prompts for common workflows:

• site-audit — Full AI-readiness audit: scores, critical issues, and prioritized action plan
• fix-plan — Generate specific code/content fixes for top issues
• competitor-analysis — Compare your site against competitors and identify gaps

License

MIT