REST API overview

The RankPine REST API — programmatic access to keywords, articles, Search Console data, and generation, built for scripts and autonomous AI agents alike.

RankPine isn't just user-friendly — it's agent-friendly. Everything your dashboard can do with keywords, articles, and Search Console data is available over a plain REST API, and as an MCP server your AI agents can plug straight into.

Base URL: https://rankpine.com/api/v1
Format: JSON in, JSON out (snake_case fields)
Auth: API key as a bearer token
MCP endpoint: https://rankpine.com/api/mcp

Authentication

Create a key in Settings → API keys (any RankPine account). Keys look like rp_live_… and are shown once at creation — store them like passwords.

Pass the key on every request:

curl https://rankpine.com/api/v1/me \
  -H "Authorization: Bearer rp_live_your_key_here"

{
  "user": { "id": "usr_…", "email": "[email protected]" },
  "api_key": { "id": "…", "name": "Claude Code" },
  "sites": [
    {
      "id": "0d3f1c9a-…",
      "url": "https://yoursite.com",
      "name": "Your Site",
      "niche": "developer tools",
      "language": "en",
      "additional_languages": ["de", "es"],
      "geo": "global",
      "autopilot": true,
      "is_active": true,
      "created_at": "2026-05-02T09:12:00.000Z"
    }
  ],
  "subscription": { "status": "active", "site_count": 1, "interval": "month" }
}

A key can access every site its owner can access — call GET /v1/me or GET /v1/sites first to get site ids for the per-site endpoints.

Rate limits

Window	Limit
Per minute	120 requests per key
Per day	10,000 requests per key

Every response carries X-RateLimit-Limit, X-RateLimit-Remaining, and X-RateLimit-Reset (unix seconds) for the per-minute window. Exceeding a limit returns 429 with a Retry-After header. Endpoints that run or enqueue AI work (suggest, generate) have tighter per-hour caps, documented on each endpoint.

Errors

Errors are JSON with a machine-readable code:

{
  "error": {
    "code": "quota_exceeded",
    "message": "This site has used all 30 articles for the current period (resets 2026-08-01T00:00:00.000Z)."
  }
}

Status	Meaning
`400`	Malformed request — `error.details` lists the failing fields
`401`	Missing, invalid, or revoked API key
`403`	Key is valid but the action isn't allowed (quota, inactive site)
`404`	Resource doesn't exist, or belongs to another account
`429`	Rate limited — honor `Retry-After`
`5xx`	Our fault — retry with backoff

Pagination

List endpoints take ?page= (default 1) and ?per_page= (default 20, max 100) and respond with the page envelope:

{ "data": ["…"], "page": 1, "per_page": 20, "total": 143 }

Async operations

Article and keyword generation is asynchronous — the endpoint returns 202 Accepted immediately and the work runs in RankPine's pipeline (a full article takes several minutes). Poll the corresponding list/detail endpoint for progress; the response's message field tells you exactly what to poll.

Endpoints at a glance

Area	Endpoints
Account	`GET /me`
Sites	`GET /sites` · `GET /sites/{site_id}`
Keywords	`GET`, `POST /sites/{site_id}/keywords` · `POST …/keywords/generate` · `POST …/keywords/suggest` · `POST …/keywords/bulk-delete` · `DELETE …/keywords/{id}`
Articles	`GET /sites/{site_id}/articles` · `GET …/articles/{id}` · `POST …/articles/generate`
Search Console	`GET /sites/{site_id}/gsc` · `POST …/gsc/sync` · `GET …/gsc/performance` · `GET …/gsc/top-queries`
Billing	`GET /subscription`

Building with an AI agent? Skip the HTTP plumbing entirely and point it at the MCP server — the same capabilities as native tools.