API for LLM Source Verification

A developer-first API that scores the reliability of sources in LLM-generated responses. Integrate trust scoring into your language model applications with a simple REST API.

Real-time ScoringMulti-source AnalysisAudit Trails

Start Free Trial View Documentation

Trust Scores

Live

UNAIDS Uganda Profile

Q:95R:90P:91

NASA Climate Data

Q:90R:85P:86

Blog Post - TechCrunch

Q:70R:85P:37

Wikipedia Article

Q:60R:40P:35

Avg Score72

Quick Start Guide

Get started with Ancestor in three simple steps

Get API Key

Bash

curl -X POST https://api.ancestor.cloud/v1/auth/keys \
  -H "Content-Type: application/json" \
  -d '{"email": "your@email.com"}'

Evaluate Sources

Bash

curl -X POST https://api.ancestor.cloud/v1/evaluate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "What is Uganda'''s HIV prevalence?",
    "answer": "According to UNAIDS, Uganda'''s HIV prevalence is 5.7% as of 2022",
    "citations": [
      {
        "url": "https://www.unaids.org/en/regionscountries/countries/uganda",
        "title": "UNAIDS Uganda Country Profile"
      }
    ]
  }'

Get Results

JSON

{
  "trust_score": 87,
  "factors": {
    "source_quality": 92,
    "recency": 85,
    "provenance": 78
  },
  "explanation": "High confidence: UNAIDS is authoritative source, data is recent (2022), has stable URL + archived copy",
  "ancestry_id": "eval_abc123"
}

See example ancestry log →

Contact for API Key

API Reference

Complete API documentation with interactive examples and detailed schemas.

POST/v1/evaluate

Evaluate source quality for LLM-generated responses with detailed trust scoring and analysis.

GET/v1/ancestry/{id}

Retrieve detailed audit trail and evaluation history for a specific ancestry ID.

Error Handling

The API uses conventional HTTP response codes to indicate success or failure. Error responses include detailed information to help debug issues.

Error Response Format

JSON

{
  "error": {
    "code": "invalid_request",
    "message": "Missing required field: question",
    "details": {
      "field": "question",
      "expected": "string",
      "received": null
    }
  },
  "request_id": "req_abc123"
}

Common Error Examples

400 Bad Request

JSON

{
  "error": {
    "code": "invalid_request",
    "message": "Invalid URL format in citations",
    "details": {
      "field": "citations[0].url",
      "value": "not-a-valid-url"
    }
  }
}

401 Unauthorized

JSON

{
  "error": {
    "code": "unauthorized",
    "message": "Invalid API key",
    "details": {
      "hint": "Check your API key in the Authorization header"
    }
  }
}

429 Too Many Requests

JSON

{
  "error": {
    "code": "rate_limit_exceeded",
    "message": "Rate limit exceeded",
    "details": {
      "limit": 100,
      "window": "1 hour",
      "retry_after": 3600
    }
  }
}

500 Internal Server Error

JSON

{
  "error": {
    "code": "internal_error",
    "message": "An internal error occurred",
    "details": {
      "request_id": "req_abc123",
      "support": "Contact support with this request ID"
    }
  }
}

HTTP Status Codes

Status Code	Description	Common Causes
200	Success	Request completed successfully
400	Bad Request	Invalid JSON, missing fields, malformed URLs
401	Unauthorized	Missing or invalid API key
403	Forbidden	API key lacks required permissions
404	Not Found	Ancestry ID not found, invalid endpoint
422	Unprocessable Entity	Valid JSON but business logic errors
429	Too Many Requests	Rate limit exceeded for your plan
500	Internal Server Error	Server error, contact support
503	Service Unavailable	Temporary maintenance or overload

Authentication

All API requests require authentication using your API key in the Authorization header:

HTTP

Authorization: Bearer YOUR_API_KEY

Rate Limits

Rate limits are enforced per API key and reset at the beginning of each time window. Exceeding limits returns a 429 status code.

Free Tier

100 requests/hour

10 requests/minute

Burst: 20 requests

Pro Plan

1,000 requests/hour

100 requests/minute

Burst: 200 requests

Enterprise

Custom limits

Dedicated infrastructure

SLA guarantees

Tip: Use the X-RateLimit-* headers in responses to monitor your usage and implement proper backoff strategies.

Webhooks

Webhooks allow you to receive real-time notifications when evaluations are completed, enabling asynchronous processing for high-volume applications.

Webhook Events

evaluation.completed

Triggered when an evaluation request is successfully processed

evaluation.failed

Triggered when an evaluation request fails after retries

Webhook Payload Example

JSON

{
  "event": "evaluation.completed",
  "timestamp": "2024-01-15T10:30:00Z",
  "data": {
    "ancestry_id": "eval_abc123",
    "request_id": "req_xyz789",
    "trust_score": 87,
    "factors": {
      "source_quality": 92,
      "recency": 85,
      "provenance": 78
    },
    "processing_time_ms": 1250,
    "webhook_id": "wh_def456"
  },
  "signature": "sha256=a1b2c3d4e5f6..."
}

Setting Up Webhooks

Bash

# Create a webhook endpoint
curl -X POST https://api.ancestor.cloud/v1/webhooks \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://your-app.com/webhooks/ancestor",
    "events": ["evaluation.completed", "evaluation.failed"],
    "secret": "your_webhook_secret"
  }'

# Use async evaluation with webhook
curl -X POST https://api.ancestor.cloud/v1/evaluate \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "question": "What is the population of Japan?",
    "answer": "According to World Bank, Japan has 125 million people",
    "citations": [{"url": "https://data.worldbank.org/country/japan"}],
    "options": {
      "async": true,
      "webhook_id": "wh_def456"
    }
  }'

Webhook Security

All webhook payloads are signed using HMAC-SHA256. Verify the signature to ensure authenticity:

JavaScript

// Node.js example
const crypto = require('crypto');

function verifyWebhookSignature(payload, signature, secret) {
  const expectedSignature = crypto
    .createHmac('sha256', secret)
    .update(payload, 'utf8')
    .digest('hex');
  
  return `sha256=${expectedSignature}` === signature;
}

// Verify in your webhook handler
const isValid = verifyWebhookSignature(
  req.body, 
  req.headers['x-ancestor-signature'], 
  process.env.WEBHOOK_SECRET
);

Note: Webhook functionality is available for Pro and Enterprise plans.

SDKs & Libraries

Official SDKs and code examples to integrate Ancestor into your applications quickly.

PythonPython 3.10+

# Python SDK
pip install ancestor-python

from ancestor import Client
client = Client(api_key="your_key")

result = client.evaluate(
    question="What is the global temperature increase?",
    answer="According to NASA, global temperature has increased by 1.1°C since 1880",
    citations=[{"url": "https://climate.nasa.gov/evidence/"}]
)

print(f"Trust Score: {result.trust_score}")
print(f"Explanation: {result.explanation}")

# Access detailed factors
factors = result.factors
print(f"Source Quality: {factors.source_quality}")
print(f"Recency: {factors.recency}")
print(f"Provenance: {factors.provenance}")

Need help getting started? Check out our comprehensive documentation and examples.

View Full Documentation

Use Cases

Integrate source quality verification across your AI-powered applications and workflows.

LLM Applications

Verify sources in language model responses to ensure your AI applications provide trustworthy information to users.

• Real-time source validation for AI assistants
• Trust scoring for generated content
• Automated fact-checking workflows

RAG Systems

Score reliability of retrieved documents in RAG pipelines to improve the quality of context-aware responses.

• Document quality assessment
• Source ranking for retrieval
• Context reliability scoring

Conversational AI

Validate citations in chatbot responses to build user trust and ensure accurate information delivery.

• Citation validation for chatbots
• Trust indicators in conversations
• Source transparency features

Research Platforms

Assess source credibility in LLM-assisted research to maintain academic and professional standards.

• Academic source verification
• Research quality assurance
• Citation credibility analysis

Ready to enhance your application with source quality verification?

Contact for Early Access

Simple, transparent pricing

Choose the plan that fits your needs. Start free and scale as you grow.

Free

$0/month

1,000 evaluations/month

API access
Basic support
100 requests/hour
Community support

Start Free

Pro

$29/month

10,000 evaluations/month

Everything in Free
Advanced analytics & insights
Priority email support
1,000 requests/hour
Webhook support
Detailed audit trails

Enterprise

Custom

Volume pricing

Everything in Pro
Custom rate limits
Dedicated support
SLA guarantees
Custom integrations

Contact Sales

All plans include comprehensive API documentation, SDKs, and 99.9% uptime SLA.

Schedule a Demo View API Documentation