API Documentation

Analyze an AI agent output and return a complete trust verification report.

Request Body

{
  "text": "string (required) — The AI-generated output to verify",
  "includeHistory": "boolean (optional, default: false)"
}

Response

{
  "trustScore": 67,
  "tier": "medium",
  "issues": [{
    "type": "citation",
    "severity": "medium",
    "title": "Uncited Statistical Claim",
    "description": "The text makes a statistical claim without source attribution.",
    "suggestion": "Add a link to the study or data source.",
    "location": { "start": 142, "end": 198, "text": "studies show 73%..." },
    "penalty": 15
  }],
  "citations": [
    { "url": "https://arxiv.org/abs/2301.04132", "status": "valid", "reputation": "trusted" }
  ],
  "confidence": { "stated": "high", "actual": "medium", "calibrated": false },
  "report": "═══════════════════════════════════════..."
}

Example

curl -X POST https://api.trustverifier.ai/v1/verify \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"text": "According to our research, GPT-5 was released in 2024..."}'

Verify multiple outputs in a single API call. Maximum 50 texts per request.

Request Body

{
  "texts": "string[] (required) — Array of AI outputs to verify"
}

Response

{
  "results": [
    { "trustScore": 85, "tier": "high", "issues": [] },
    { "trustScore": 42, "tier": "low", "issues": [...] }
  ]
}

Retrieve a previously stored verification result by ID.

curl https://api.trustverifier.ai/v1/verify/vrf_abc123 \
  -H "Authorization: Bearer YOUR_API_KEY"