TypeSafe logo

TypeSafe API · Ai Data

Jev score -- position on a rubric with the TypeSafe API

Judges state against one score question (Jev's "score" type) and returns a probability-weighted position along an ordered rubric (can land between levels, e.g. You call it with one looot token, no TypeSafe account needed. Each call costs $0.00033 per call. Over the last 30 days looot ran it 63 times: 100% finished without an error, median answer 228 ms.

Price
$0.00033 per call
Success rate, 30 days
100%
Runs, 30 days
63
Access
looot token only

Jev score -- position on a rubric: call it from code

One POST to /v1/runs with your looot token, the endpoint id typesafe-score and an input. Values come from the stored input schema; a value in <angle brackets> is a placeholder the schema gives no example for.

InputTypeValue shown
statestringplaceholder
questionsobjectplaceholder

POST /v1/runs

TypeSafe API request: curl
curl -X POST 'https://api.looot.ai/v1/runs?wait=true' \
  -H 'Authorization: Bearer YOUR_LOOOT_TOKEN' \
  -H 'Content-Type: application/json' \
  -d '{"endpointId":"typesafe-score","input":{"state":"<string>","questions":"<object>"},"idempotencyKey":"<unique-idempotency-key>"}'

Response: verified on real answers

{ model, usage: { input_tokens, output_tokens }, answers: { frustration: { type, score, legend: { ... }, confidence, probabilities: { ... } } } }

validated on 2 real answers.

Price and reliability

Price formula
$0.00033 per call
Most one run can cost
$0.00033
Success rate, 30 days
100%
Runs, 30 days
63
Median answer time
228 ms
Last success
Oct 30, 2026

Run numbers are looot's own runs of each endpoint over the last 30 days: the share that completed without an error, and the median time to answer. An endpoint with fewer than 10 runs in that window shows "not enough runs to rank" and is never ranked. Prices are what looot charges per call or per result, fee included, read from the live catalog; you prepay credit and each call is charged when it runs.

More TypeSafe endpoints

Questions about the TypeSafe API on looot

Judges state against one score question (Jev's "score" type) and returns a probability-weighted position along an ordered rubric (can land between levels, e.g. 1.6), with the legend and a confidence -- no generated text. Exactly one question per call, 2-10 rubric levels written as concrete situations (not adjectives). looot stores its output format, checked on real answers.

It costs $0.00033 per call. You keep a prepaid balance. The price is shown before each call runs, and you are charged only for calls that actually ran.

Send POST https://api.looot.ai/v1/runs with your looot token in the Authorization header and a body with endpointId "typesafe-score" and an input object (state, questions). The looot CLI runs it with looot run typesafe-score, and the looot MCP server with its run tool.

Whether a miss costs money depends on the provider. Some bill only for found values. Some bill every call. A failed call costs nothing. You can request a refund of the unused part of the balance you paid for at any time.

Over the last 30 days looot ran it 63 times: 100% finished without an error, median answer 228 ms.

Give your agent every data provider, one key.

Everyone gets $2 in free credits once after sign-in and workspace selection. Then pay only for what you use.