---
title: "Queue a rank check"
method: POST
path: "/keywords/{keyword_id}/checks"
tags: ["rank-checks"]
---

# Queue a rank check

`POST /keywords/{keyword_id}/checks`

## Request

- Base URL: `https://eu.bisibility.com/api/v1`
- URL: `https://eu.bisibility.com/api/v1/keywords/{keyword_id}/checks`
- Auth: one of:
  - HTTP bearer
  - HTTP bearer
  - OAuth 2 (authorization code, authorization URL `/api/auth/oauth2/authorize`, token URL `/api/auth/oauth2/token`, scopes: `api:write`)

Queues a one-item rank-check run unless inline execution is configured.

## Path parameters

- `keyword_id` string, required: Public keyword ID.

## Query parameters

- `project` string

## Headers

- `Bisibility-API-Version` 'v1'
- `Idempotency-Key` string
- `X-Bisibility-Project` string
- `X-Bisibility-Source` 'sdk' | 'cli' | 'mcp'

## Request body

- object
  - `provider_id` string
  - `max_cost_cents` integer: Best-effort pre-estimate provider cost gate, in cents. The launch is refused with cost_limit_exceeded when its preflight estimate exceeds it.

## Response `201`

Rank check completed when inline execution is enabled

- RankCheck
  - `usage` object: Confirmed provider usage; null quantity means unconfirmed and is never inferred from requested depth. Costs may take up to 15 minutes to update.
    - `quantity` number, nullable, required
    - `status` 'confirmed' | 'unconfirmed', required
    - `unit` 'operations' | 'cents', required
  - `attempts` object[], nullable, required: Provider fallback attempts recorded before the final rank-check status.
    - `message` string, required
    - `provider` string, required
  - `checked_at` string, date-time, required
  - `cost_cents` number, nullable, required
  - `error` string, nullable, required
  - `error_code` string, nullable, required: Stable failure code for this check, or null for rows older than the code taxonomy.
  - `id` string, required
  - `keyword_id` string, required
  - `observation_completeness` 'complete' | 'truncated_by_stop_on_match' | 'unknown' | 'null', nullable, required: Persisted coverage, or null when legacy metadata is absent. Only complete coverage supports absence within depth; a valid positive rank remains valid with truncated coverage.
  - `position` integer, nullable, required
  - `previous_position` integer, nullable, required
  - `provider` string, required
  - `ranking_url` string, nullable, required
  - `run` object, nullable, required: The run that produced this result, or null for a legacy row.
    - `finished_at` string, date-time, nullable, required
    - `id` string, required
    - `started_at` string, date-time, nullable, required
    - `status` string, required
    - `trigger` string, required
  - `run_id` string, nullable, required: Rank-check run that produced this result, or null for a legacy row.
  - `status` 'completed' | 'failed' | 'running', required

## Other responses

- `202`: Rank check queued
- `400`: Bad request
- `401`: Unauthorized
- `402`: Deployment credits exhausted for a paid provider request
- `403`: Forbidden, or `byok_required`: the account runs paid provider work only on its own provider keys and none serves this request
- `404`: Not found
- `409`: Unsupported API version
- `422`: The launch estimate exceeds max_cost_cents
- `429`: Rate limited

---

[API](https://skmtc.dev/bisibility/apis/bisibility-public-api.md) · [All operations](https://skmtc.dev/bisibility/apis/bisibility-public-api/llms.txt) · [OpenAPI document](https://skmtc.dev/bisibility/apis/bisibility-public-api/revisions/a6e1ec0dec82?raw)