Run exactly the next framework stage for a decision
The recommended way to run analysis: call this repeatedly (checking is_complete) to work through a decision one stage at a time. The stage index and prior-stage context are derived server-side — no request body needed. Prefer this over /analyze, which runs every stage in one request and can silently exceed the request time limit on longer frameworks.
/v1/decisions/{id}/advanceAPI key in dc_live_... / dc_test_... format, from /settings/api-keys.
In: header
Path Parameters
Decision or job ID.
Response Body
curl -X POST "https://api.deecee.ai/v1/decisions/string/advance"{
"success": true,
"data": {
"stage_name": "string",
"stage_content": "string",
"stage_index": 0,
"total_stages": 0,
"is_complete": true,
"tokens_used": 0,
"job_id": "string"
}
}{
"success": true,
"data": {
"stage_name": "string",
"stage_content": "string",
"stage_index": 0,
"total_stages": 0,
"is_complete": true,
"tokens_used": 0,
"job_id": "string"
}
}{
"success": false,
"error": "string",
"code": "string"
}{
"success": false,
"error": "string",
"code": "string"
}{
"success": false,
"error": "string",
"code": "string"
}API Reference
REST endpoints for the DeeCee.ai agentic API.
Commit a decision's title/framework/complexity
Typically the recommendation from POST /v1/decisions/{id}/assess, as-is or with overrides — only call this after the user has confirmed it. Synchronous, no LLM call. Once applied, ask the user whether they want stage-by-stage (/advance) or all-at-once (/analyze) analysis before starting it.