MetaMonster API
The MetaMonster API lets you read and manage your SEO data programmatically: sites, pages, content, snapshots, analysis, drafts, and keyword metrics. It's the same data you see in the MetaMonster dashboard, available over HTTP.
Base URL
https://new.metamonster.ai/api/v1
Every endpoint in this documentation is relative to that base. All requests use HTTPS.
Quick start
curl https://new.metamonster.ai/api/v1/me \
-H "Authorization: Bearer mm_YOUR_API_KEY"
If your key is valid, you'll get back its name and scopes:
{ "name": "Production integration", "prefix": "mm_ab12", "scopes": ["sites:read", "content:write", "cms:publish"] }
New here? Start with the Getting started guide. It walks you from zero to your first authenticated call.
Documentation
| Guide | What's in it |
|---|---|
| Getting started | Create an API key and make your first call |
| Authentication | Key format, the Authorization header, scopes, key lifecycle |
| Conventions | Response envelope, pagination, errors, rate limits, versioning |
| MCP server | The remote MCP server for chat clients (Claude Code, Cursor, Claude Desktop): setup, the tool list, and URL resolution |
Endpoint reference
| Resource | Endpoints |
|---|---|
| Health & identity | GET /health, GET /openapi.yaml, GET /me |
| Sites | List and read your sites |
| Site planning | Plan a site that isn't live yet: propose its pages from a brief, assign keywords, then go live |
| Opportunities | The ranked "what should I work on" report for a site |
| Rubric | The components, weights, criteria, and grade bands the LLM grader scores a site's pages against |
| Site facts | Read and write the verified business facts (identity, contact, testimonials, stats, awards, clients, people) that content generation treats as ground truth |
| Pages | Read pages, update metadata (one page or up to 50 at once), read and write body content, trigger analysis (single or bulk) and recrawl, read an analysis, compare it against the previous run, and read score history |
| Checks | Free deterministic hygiene checks on a page's current state (these are not the grade) |
| Brief | The page work packet: fields, analysis, recommendations, keywords, GSC, outline, and links in one call |
| Links | Outbound links extracted from a page's latest crawl |
| Outlines | A page's latest content outline and the SERP snapshot it was generated from |
| Search Console data | Live Google Search Console performance for a site or page |
| Snapshots | Point-in-time captures of a page |
| Content versions | Version history for a page's body content, and how to roll back |
| Drafts | Read and write field drafts (title, meta description, etc.) |
| Recommendations | Dismiss, resolve (with AI verification), and recheck a page's recommendations |
| Jobs | Poll the status of an async analysis or recrawl trigger, one job at a time or up to 50 in a single request |
| Keyword metrics | Look up search volume, difficulty, and intent for a keyword |
For AI agents
If you're driving the API with an AI coding agent (Claude Code, Cursor, etc.), see skills/metamonster-api/: a self-contained skill that teaches an agent the auth flow, the endpoints, and the common workflows.
A machine-readable OpenAPI 3.1 spec is also available for generating SDKs, importing into Postman, or rendering with Swagger UI. It's served live at https://new.metamonster.ai/api/v1/openapi.yaml with no authentication required, so tools and agents can fetch it directly.
Core concepts
- Sites are the websites you've added to MetaMonster. Everything else hangs off a site.
- Pages are the individual URLs discovered on a site. Pages are long-lived.
- Snapshots are point-in-time captures of a page's crawled content (title, meta description, body, etc.).
- Analysis is MetaMonster's SEO scoring and recommendations for a page.
- Drafts are proposed new values for a single page field (like a rewritten
titleormeta_description) that you can create, edit, and later publish.
Getting help
Questions or something not working as documented? Reach out to the MetaMonster team.