Generate Prompt Suggestions
Changed onGenerate Prompt Builder suggestions for a project, the same way the Prompt Builder does. Generation runs in the background: this returns its generation ID as soon as it is queued. Poll GET /prompts/suggestions/generations/{generation_id} until it succeeds or fails, then list successful results with GET /prompts/suggestions?legacy=false. topic_id, prompt_suggestion_id and segment each narrow the run and are mutually exclusive; with an empty body every topic awaiting review is generated for again, replacing its current suggestions. The project's branding/intent split has to be configured first, with PUT /project-profile — generation has nothing to run against without it. Markets are optional: without them the project's own country and language are used. A project gets one full run of suggestions a day, sized as every topic, market, branding and intent combination at 10 prompts each. Once the project has queued that many, the request returns 429 with the time to retry.
Request
- Base URLhttps://api.peec.ai/customer/v1
- URLhttps://api.peec.ai/customer/v1/prompts/suggestions/generate
- Auth
- API key in header X-API-Key
- or API key in query parameter api_key
- or HTTP bearer
Request body
Example request
{
"topic_id": "to_e6b8cdd3-a51b-4d94-a866-28dbe6b830a6",
"prompt_suggestion_id": "pr_93f790de-5b7a-45ee-b782-61103c81f20d"
}Response
Generation queued
Changes
- info
endpoint added
- info