Seed a decision with labeled examples

Imports labeled starting examples (seeds): each runs through the engines (spends engine calls) and is stored as an evaluation with seed feedback. Fresh responses include recorded cost per result and a batch total; engine_calls counts upstream calls, which can answer several features, and cache_hits is 0 today. Evaluates up to 200 labeled examples and records their claims as seed feedback in one atomic batch, so previews can replay them. Before any engine call each example is checked: media and turn parts are 400 CAPABILITY_UNSUPPORTED (param examples.<i>.subject.<j>.type), an inputs name its subject lacks is 400 CONTENT_INVALID (examples.<i>.nodes.<leaf>.inputs.<k>), and claims are checked (examples.<i>.claims.<j>). If any example’s output fails, the whole batch is refused with 502 engine_failed and details.items naming every failed example with its per-feature failures (features); nothing is saved. The request has a 60 s budget, the commit included; a batch that cannot be evaluated within it is a 422 naming how many examples were evaluated in time (send smaller batches). A replay with the same Idempotency-Key returns the first result with the Idempotent-Replayed header and calls no engine.

POST /v1/decisions/{decision_id}/seeds

Seed a decision with labeled examples

cURL

curl --request POST \
  --url https://api.moonbounce.ai/v1/decisions/{decision_id}/seeds \
  --header 'Authorization: Bearer <token>' \
  --header 'Content-Type: application/json' \
  --data '
{
  "spec": {
    "format": "moonbounce.spec/0",
    "nodes": {},
    "output": "<string>",
    "guidance": "<string>",
    "target": "<string>",
    "meta": {
      "title": "<string>",
      "description": "<string>",
      "owner": "<string>",
      "tags": [
        "<string>"
      ],
      "display": {}
    }
  },
  "examples": [
    {
      "item_id": "<string>",
      "subject": [
        {
          "type": "text",
          "text": "<string>",
          "name": "<string>",
          "sensitive": true
        }
      ],
      "claims": [
        {
          "kind": "target",
          "value": true
        }
      ],
      "context": [
        {
          "type": "text",
          "text": "<string>",
          "name": "<string>",
          "sensitive": true
        }
      ],
      "summary": "<string>",
      "metadata": {},
      "reason": "<string>",
      "unsure": true
    }
  ],
  "weights": "default",
  "structure": "<string>"
}
'

201

{
  "environment": "production",
  "decision_id": "<string>",
  "version": {
    "structure": "<string>",
    "weights": "<string>"
  },
  "total": 123,
  "agreed": 123,
  "results": [
    {
      "item_id": "<string>",
      "evaluation_id": "<string>",
      "feedback_id": "<string>",
      "status": "scored",
      "agreed": true,
      "conflicts": [
        {
          "feature": "<string>",
          "expected": true,
          "got": 0.5
        }
      ],
      "outcome": "<string>",
      "score": 123,
      "cost": {
        "engine_calls": 1,
        "cache_hits": 1,
        "input_tokens": 1,
        "usd_micros": 1
      }
    }
  ],
  "cost": {
    "engine_calls": 1,
    "cache_hits": 1,
    "input_tokens": 1,
    "usd_micros": 1
  }
}

Authorizations

Authorization · string · header · required

An API key, sent as Authorization: Bearer <key>. Each operation lists the scope its key needs: decisions:evaluate (evaluate and read evaluations), decisions:feedback (your app's feedback) or decisions:develop (everything else, including owner feedback and the review queue). An operation listing no scope takes any scoped key. A key with no scope, or without the operation's, gets 403 INSUFFICIENT_SCOPE.

Headers

idempotency-key · string

Makes a retry safe. For 24 hours, the same key with the same request returns the first stored result (with Idempotent-Replayed: true); the same key with a different request is a 422 IDEMPOTENCY_KEY_REUSED. Keys are scoped to the organization and environment across every operation, so use a fresh key per request. A request that failed before its result was stored runs again when retried with its key.

Required string length: 1 - 255

Pattern: ^[\x20-\x7e]+$

Path Parameters

decision_id · string · required

Pattern: ^[a-z][a-z0-9_.-]{0,127}$

Body

application/json

Option 1

Option 2

spec · object · required

Show child attributes

examples · object[] · required

Required array length: 1 - 200 elements

Show child attributes

weights · enum<string> string · default: default

Available options: default

outcomes · object

Show child attributes

structure · string

Pattern: ^sha256:[0-9a-f]{64}$

Response

201

application/json

Seeded

environment · enum<string> · required

The environment of the key that made the call: production or test. Evaluations, feedback, reviews and suggestions are separate per environment; registered structures are shared.

Available options: production, test

decision_id · string · required

Pattern: ^[a-z][a-z0-9_.-]{0,127}$

version · object · required

Show child attributes

total · integer · required

agreed · integer · required

Results whose agreed value is true.

results · object[] · required

Show child attributes

cost · object

Always present on fresh responses; absent only on an idempotent replay of a response stored before costs were reported.

Show child attributes