GET /v1/prompts/publish

Summary

Search the catalogue of published prompts by keyword. Results are ranked by relevance using a combination of text similarity and semantic matching. Video prompts are excluded from results.

HTTP Request

GET /v1/prompts/publish?q=<query>

Query parameters

Parameter	Type	Required	Description
q	string	Yes	Search query. Maximum 200 characters; leading/trailing whitespace is trimmed.

Response

Success (200 OK)

Returns a JSON object whose status is "SUCCESS" and a data object containing:

Field	Type	Description
rows	array of objects	List of matching prompts, ranked by relevance.
rows[].promptId	string	Identifier of the published prompt.
rows[].promptVersionId	string	Identifier of the latest published version.
rows[].title	string	Title of the prompt.
rows[].description	string	Short description of the prompt.
rows[].slug	string	URL-friendly slug for the prompt.
rows[].model	integer	Model index used by the prompt. See Models & Options.
rows[].modality	integer	0 for text, 1 for image.
rows[].estimatedCredits	integer	Estimated credit cost per run.
rows[].uses	integer	Total number of times the prompt has been run.
rows[].upVotes	integer	Number of upvotes.
rows[].publishedAt	string (ISO 8601)	When the latest version was published.
rows[].samples	object	Sample inputs/outputs for the prompt.
rows[].configs	object	Prompt configuration metadata.

Example response

{
  "status": "SUCCESS",
  "data": {
    "rows": [
      {
        "promptId": "01234567-89ab-cdef-0123-456789abcdef",
        "promptVersionId": "a1b2c3d4-5678-90ab-cdef-1234567890ab",
        "title": "Blog post writer",
        "description": "Generate a blog post from a topic and tone.",
        "slug": "blog-post-writer",
        "model": 50,
        "modality": 0,
        "estimatedCredits": 150,
        "uses": 2340,
        "upVotes": 187,
        "publishedAt": "2025-11-03T10:15:30.000Z",
        "samples": {},
        "configs": {}
      }
    ]
  }
}

Error responses

HTTP status	status code	Description
422	VALIDATION_ERROR	Missing or invalid q parameter.

See also

• Retrieve a prompt: GET /v1/prompts/:promptId
• Run a prompt: POST /v1/runs