AI

Run Ai Column

Queue the column's cells for the given rows (RO-340); a background job fills them through the provider layer, one small AI call per row, charged as AI credits (task `ai_column`). Cells with a result are kept unless `force`. Show the cost first with `POST /ai-credits/estimate` (`action` `ai_column`, `items`). 429 `ai_credits_exhausted` / `ai_credits_user_limit` at the monthly allowance or the member's limit; 429 `ai_columns_daily_limit` past 500 cells a day per workspace.

API key access

Needs the leads:write scope with an API key (ro_pat_...). A signed-in session uses the caller's role.
POST/api/v1/ai-columns/{column_id}/run
AuthorizationBearer <token>

Access token from /api/v1/auth/login, or an API key (ro_pat_..., RO-176) from /api/v1/workspaces/current/api-keys: the key implies its workspace and needs the route's scope. MCP connectors (/api/v1/mcp only) may use an OAuth access token (ro_oat_..., RO-284).

In: header

Path Parameters

column_idstring

The column.

Formatuuid

Header Parameters

X-Workspace-Id?string | null

Id of the workspace to act on (must be a member).

idsarray<string>

The leads or companies to run it on (1-100).

Items1 <= items <= 100
force?boolean | null

Run cells that already have a result again (default false: they are kept).

Response Body

curl -X POST "http://localhost:8000/api/v1/ai-columns/497f6eca-6276-4993-bfeb-53cbbbba6f08/run" \  -H "X-Workspace-Id: string" \  -H "Content-Type: application/json" \  -d '{    "ids": [      "497f6eca-6276-4993-bfeb-53cbbbba6f08"    ]  }'
{
  "column_id": "8f0515e3-00f6-4dd0-8153-321f1f40863d",
  "queued": 0,
  "cached": 0,
  "missing": [
    "497f6eca-6276-4993-bfeb-53cbbbba6f08"
  ]
}
Empty
{
  "detail": [
    {
      "loc": [
        "string"
      ],
      "msg": "string",
      "type": "string",
      "input": null,
      "ctx": {}
    }
  ]
}
Empty