journeyapi.dev

POST/v1/prompts/analyze

Breaks a prompt into terms, reports their relative influence, and proposes shorter rewrites that keep the parts doing the work. Long prompts often dilute themselves; this shows where.

Body

promptstringrequired: The prompt to analyse.
modelstring: image-2 (default), image-2-lite for a cheaper faster render, or illustration-1 for drawn and stylised work.
webhook_urlstring: Overrides the account webhook for this job only.
metadataobject: Up to 16 key-value pairs echoed back on the job, for correlating with your own records.

Request

curl -X POST https://api.journeyapi.dev/v1/prompts/analyze \
  -H "Authorization: Bearer $JOURNEY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "prompt": "a very beautiful highly detailed lighthouse on a basalt cliff in heavy fog, 8k, masterpiece" }'

Response · 200 OK, once complete

{
  "id": "job_01HQ8W28XR5CLF",
  "object": "job",
  "type": "prompt_analysis",
  "status": "succeeded",
  "terms": [
    { "text": "lighthouse", "weight": 0.94 },
    { "text": "basalt cliff", "weight": 0.81 },
    { "text": "heavy fog", "weight": 0.66 },
    { "text": "highly detailed", "weight": 0.11 },
    { "text": "8k", "weight": 0.04 },
    { "text": "masterpiece", "weight": 0.02 }
  ],
  "suggestions": [
    "lighthouse on a basalt cliff, heavy fog",
    "lighthouse, basalt cliff, fog",
    "a lighthouse on a basalt cliff in heavy fog"
  ],
  "cost_usd": 0.01,
  "created_at": "2026-09-18T09:18:44Z",
  "completed_at": "2026-09-18T09:18:46Z"
}

Analyse a prompt

Body