Analyze API

The analyze API is the public face of Lumni’s detectors. It’s the same trace-only detection engine that runs on production traffic, but with no account and no integration — paste a trace, get a teardown.

Endpoint

curl -sS https://agentf.lumniverse.com/v1/public/analyze \
  -H 'content-type: application/json' \
  -d @trace.json

No authentication. Rate-limited per IP.

Request

Send a trace as JSON. Lumni normalizes common shapes, but the model it works against is the standard run + steps structure:

{
  "userRequestText": "Refund my last order",
  "success": true,
  "outcomeSummary": "Your refund has been processed.",
  "metadata": { "model": "gpt-4o", "costUsd": 0.021 },
  "steps": [
    { "stepKind": "tool", "stepName": "issue_refund", "status": "failed", "errorMessage": "502 Bad Gateway" },
    { "stepKind": "model", "outputSummary": "Your refund has been processed." }
  ]
}

Response

Failures, most-suspicious first:

{
  "failures": [
    {
      "detectorKey": "false_success",
      "suspicionScore": 95,
      "severity": "high",
      "initialFailureSource": "tool",
      "title": "False success: agent claimed completion after issue_refund failed",
      "summary": "Step \"issue_refund\" reported \"502 Bad Gateway\", but the agent's final message claimed success (\"Your refund has been processed.\"). The user was told the action completed when it did not.",
      "primaryStepId": "step_1"
    }
  ]
}

An empty failures array means no trace-only detector fired — Lumni found nothing suspicious in what you sent.

Privacy

The playground is designed for anonymous use:

No auth, no account. Nothing is tied to you.
No PII captured. Send an anonymized trace.
Rate-limited per IP to keep it available.