What does Jev AI Agent Trace Evaluator — Score LLM Traces return?

It returns 12 documented output fields: state, questions, index, sourceId, trace, evidenceSupport, toolRelevance, humanReviewProbability, status, minConfidence, judgments and model. Each field declares its type, a description of what it holds, and an example value.

What input does Jev AI Agent Trace Evaluator — Score LLM Traces require?

records is required. A further 2 inputs are optional. The full input schema, with types and defaults, is below.

Does Jev AI Agent Trace Evaluator — Score LLM Traces need an API key or login?

Yes — typesafeApiKey is marked secret and must be supplied by you. Nothing else in the input schema needs credentials.

How do I run Jev AI Agent Trace Evaluator — Score LLM Traces?

It runs on Apify. Start it from its listing with the inputs above, or call the Apify API with the Actor's ID. Results land in a dataset you can export as JSON, CSV or Excel, or read through the Apify API into your own pipeline.

Jev AI Agent Trace Evaluator — Score LLM Traces

Evaluate supplied AI-agent traces for evidence support, tool relevance, and review need using a TypeSafe Jev key. It never executes tools.

Jev AI Agent Trace Evaluator — Score LLM Traces returns 12 documented output fields, including state, questions, index, sourceId and trace. Every field carries a declared type, a written description and an example value. 18 runs are recorded on Apify.

Run it on ApifyPay per result · no subscription

12: Output fields
3: Inputs
18: Runs
2026-10-15: Last run

01 — output

What it returns

Every field the extractor writes to its dataset, with its type, a written description and an example value.

Field	Type	What it holds	Example
state	stringnull	Current state of this record in its own lifecycle.
questions	objectnull	Questions as recorded for this row.
index	string \| number \| boolean \| array \| objectnull	Position of this row within the run.	1
sourceId	string \| number \| boolean \| array \| objectnull	Identifier of the input record this row came from.	rec_01
trace	string \| number \| boolean \| array \| objectnull	Trace as recorded for this row.	[]
evidenceSupport	string \| number \| boolean \| array \| objectnull	Evidence Support as recorded for this row.	[]
toolRelevance	string \| number \| boolean \| array \| objectnull	Tool Relevance as recorded for this row.	[]
humanReviewProbability	string \| number \| boolean \| array \| objectnull	Human Review Probability as recorded for this row.	[]
status	string \| number \| boolean \| array \| objectnull	Outcome recorded for this row.	ok
minConfidence	string \| number \| boolean \| array \| objectnull	Min Confidence as recorded for this row.	[]
judgments	string \| number \| boolean \| array \| objectnull	The full structured judgement returned by the model.	{"stance":{"choice":"recommended"}}
model	string \| number \| boolean \| array \| objectnull	Identifier of the model that produced the judgement.	jev-latest

02 — sample

One record

Assembled from the example values declared on the schema above, not invented at render time.

{
  "state": "",
  "questions": "",
  "index": 1,
  "sourceId": "rec_01",
  "trace": "[]",
  "evidenceSupport": "[]",
  "toolRelevance": "[]",
  "humanReviewProbability": "[]"
}

03 — input

What you give it

Input	Type	Description
records	arrayrequired	Rows with request, tool result, and final answer context.
typesafeApiKey	stringoptional	Secret — supply your own value.
maxItems	integeroptional	Hard cap on processed trace rows.

04 — run

Run Jev AI Agent Trace Evaluator on your own input

It runs in the cloud on Apify. Nothing to install, and you are billed per result returned rather than per attempt.

Open on Apify ↗

Also in ai, developer tools

05 — questions

Asked before running it

12 of 12 fields described

What does Jev AI Agent Trace Evaluator — Score LLM Traces return?: It returns 12 documented output fields: state, questions, index, sourceId, trace, evidenceSupport, toolRelevance, humanReviewProbability, status, minConfidence, judgments and model. Each field declares its type, a description of what it holds, and an example value.
What input does Jev AI Agent Trace Evaluator — Score LLM Traces require?: records is required. A further 2 inputs are optional. The full input schema, with types and defaults, is below.
Does Jev AI Agent Trace Evaluator — Score LLM Traces need an API key or login?: Yes — typesafeApiKey is marked secret and must be supplied by you. Nothing else in the input schema needs credentials.
How do I run Jev AI Agent Trace Evaluator — Score LLM Traces?: It runs on Apify. Start it from its listing with the inputs above, or call the Apify API with the Actor's ID. Results land in a dataset you can export as JSON, CSV or Excel, or read through the Apify API into your own pipeline.