Drazill API Documentation

{
  "brier_score": "string",
  "computed_at": "2024-01-15T09:30:00Z",
  "percentile": "string",
  "rank": 1,
  "total_ranked_users": 1,
  "total_resolved": 1,
  "user_id": "string",
  "methodology": {
    "formula_version": "string",
    "confidence_level": "LOW",
    "data_window": "string",
    "informational_only": true,
    "last_computed_at": "2024-01-15T09:30:00Z",
    "limitations": [
      "string"
    ],
    "sample_size": 1
  },
  "skill": {
    "brier": 1.1,
    "composite_skill": 1.1,
    "computed_at": "2024-01-15T09:30:00Z",
    "ece": 1.1,
    "log_loss": 1.1,
    "model_version": "string",
    "n_resolved": 0,
    "price_discovery_bps": 1.1,
    "provisional": true,
    "sharpness": 1.1,
    "skill_pvalue": 1.1
  }
}

Authentication

AuthorizationBearer

Bearer authentication of the form Bearer <token>, where token is your auth token.

Response

Successful Response

brier_scorestringformat: "^(?!^[-+.]*$)[+-]?0*\d*\.?\d*$"

computed_atdatetime

percentilestringformat: "^(?!^[-+.]*$)[+-]?0*\d*\.?\d*$"

0-100, higher = better predictor

rankinteger

total_ranked_usersinteger

total_resolvedinteger

user_idstringformat: "uuid"

methodologyobject or null

Explainable methodology metadata for trust and intelligence metrics.

skillobject or null

ML-10 skill & calibration suite for a user (additive; shadow-mode).

Every field is optional so it can ride alongside the existing predictor score without ever changing it. Populated only when ML_SKILL_SUITE_ENABLED and the user has an ml_user_skill_scores row.

Errors

401

Unauthorized Error

403

Forbidden Error

422

Unprocessable Entity Error

429

Too Many Requests Error

500

Internal Server Error

1	from drazill import DrazillClient
2
3	client = DrazillClient(api_key="drzl_test_...", base_url="https://api.drazill.com/api/v1")
4
5	result = client.intelligence.get_predictor_score()
6	print(result)

Get your Drazill Predictor Score

Authentication

Headers

Response

Errors