generated: '2026-07-18' method: searched source: https://www.bolna.ai/docs/api-reference docs: api_reference: https://www.bolna.ai/docs/api-reference/introduction pagination: https://www.bolna.ai/docs/api-reference/pagination rate_limiting: https://www.bolna.ai/docs/api-reference/rate-limiting limits: https://www.bolna.ai/docs/api-reference/limits errors: https://www.bolna.ai/docs/api-reference/errors authentication: style: bearer-api-key header: 'Authorization: Bearer ' detail: See authentication/bolna-authentication.yml idempotency: supported: false notes: >- No idempotency-key header or parameter is documented. Call placement is inherently side-effecting; de-duplication is the caller's responsibility. pagination: style: offset mechanism: page-number params: page_number: { type: integer, default: 1, description: First page starts at 1 } page_size: { type: integer, default: 20, max: 50 } response_fields: [total, page, page_size, has_more, data] notes: Offset-based under the hood; page_number/page_size are query parameters. versioning: scheme: uri-path current: v2 detail: >- Agent endpoints are versioned in the path (/v2/agent). v1 agent endpoints are marked deprecated in the API reference. See lifecycle/bolna-lifecycle.yml. error_envelope: format: custom-json shape: '{ "error": , "message": }' problem_json: false detail: >- Not RFC 9457. All 4xx/5xx responses return a JSON body with an integer `error` code and a human-readable `message`. See errors/bolna-error-codes.yml. rate_limit_signaling: scope: per-organization (or per-user when no org) default_limit: 1000 requests/minute endpoint_specific: 500 requests/minute (agent get, agent executions, /call) over_limit_status: 429 guidance: Exponential backoff and retry. detail: See rate-limits/bolna-rate-limits.yml datetime: scheduled_at_format: ISO 8601 with numeric UTC offset (e.g. 2026-06-23T18:30:00+00:00) notes: The 'Z' suffix is rejected (returns HTTP 500); use +00:00 instead. webhooks: detail: See asyncapi/bolna-webhooks.yml