Contract Analysis

Overview

This cookbook demonstrates how to analyze contracts using Aifano’s Parse and Extract endpoints. You’ll learn to extract parties, dates, key clauses, and obligations from legal documents — turning dense PDFs into structured, queryable data.

What You’ll Build

A workflow that:

Parses a contract into structured sections
Extracts key metadata (parties, dates, terms)
Identifies specific clauses (termination, liability, confidentiality)

Step 1: Parse the Contract

First, parse the contract to get the full text with section structure:

import requests

AIFANO_API_KEY = "ak_live_your_key_here"
BASE_URL = "https://platform.aifano.com"

# Parse the contract
parse_result = requests.post(
    f"{BASE_URL}/parse",
    headers={"Authorization": f"Bearer {AIFANO_API_KEY}"},
    json={"input": "aifano://service-agreement.pdf"}
).json()

# Print section headers
for chunk in parse_result["result"]["chunks"]:
    for block in chunk["blocks"]:
        if block["type"] in ("Title", "Section Header"):
            print(f"  [{block['type']}] {block['content']}")

const AIFANO_API_KEY = "ak_live_your_key_here";
const BASE_URL = "https://platform.aifano.com";

const parseResult = await fetch(`${BASE_URL}/parse`, {
  method: "POST",
  headers: {
    "Authorization": `Bearer ${AIFANO_API_KEY}`,
    "Content-Type": "application/json"
  },
  body: JSON.stringify({ input: "aifano://service-agreement.pdf" })
}).then(r => r.json());

parseResult.result.chunks.forEach(chunk => {
  chunk.blocks
    .filter(b => ["Title", "Section Header"].includes(b.type))
    .forEach(b => console.log(`  [${b.type}] ${b.content}`));
});

Step 2: Extract Key Contract Data

Use the job reference from Step 1 to extract structured data without re-parsing:

Python

schema = {
    "type": "object",
    "properties": {
        "contract_title": {"type": "string"},
        "effective_date": {"type": "string", "description": "YYYY-MM-DD format"},
        "expiration_date": {"type": "string", "description": "YYYY-MM-DD format"},
        "parties": {
            "type": "array",
            "items": {
                "type": "object",
                "properties": {
                    "name": {"type": "string"},
                    "role": {"type": "string", "description": "e.g., Client, Provider, Vendor"},
                    "address": {"type": "string"}
                }
            }
        },
        "contract_value": {"type": "number", "description": "Total contract value"},
        "currency": {"type": "string"},
        "payment_terms": {"type": "string"},
        "termination_clause": {"type": "string", "description": "Summary of termination conditions"},
        "liability_cap": {"type": "string", "description": "Maximum liability amount or terms"},
        "confidentiality": {"type": "boolean", "description": "Whether an NDA/confidentiality clause exists"},
        "governing_law": {"type": "string", "description": "Jurisdiction or governing law"},
        "renewal_terms": {"type": "string", "description": "Auto-renewal or renewal conditions"}
    }
}

# Use jobid:// to skip re-parsing (saves credits!)
job_id = parse_result["job_id"]
extract_result = requests.post(
    f"{BASE_URL}/extract",
    headers={"Authorization": f"Bearer {AIFANO_API_KEY}"},
    json={
        "input": f"jobid://{job_id}",
        "schema": schema,
        "system_prompt": "Extract all contract metadata. Summarize clauses concisely. Use YYYY-MM-DD for dates."
    }
).json()

import json
print(json.dumps(extract_result["result"], indent=2))

Step 3: Example Output

{
  "contract_title": "Master Service Agreement",
  "effective_date": "2024-01-01",
  "expiration_date": "2025-12-31",
  "parties": [
    { "name": "Acme Corp", "role": "Client", "address": "Berlin, Germany" },
    { "name": "TechServ GmbH", "role": "Provider", "address": "Munich, Germany" }
  ],
  "contract_value": 120000,
  "currency": "EUR",
  "payment_terms": "Net 30 days from invoice date",
  "termination_clause": "Either party may terminate with 90 days written notice. Immediate termination for material breach.",
  "liability_cap": "Limited to 12 months of fees paid",
  "confidentiality": true,
  "governing_law": "German law, courts of Berlin",
  "renewal_terms": "Auto-renews for 1-year periods unless terminated 90 days prior"
}

Step 4: Split Multi-Section Contracts

For contracts with multiple sections (e.g., MSA + SOW + NDA), use Split first:

Python

split_result = requests.post(
    f"{BASE_URL}/split",
    headers={"Authorization": f"Bearer {AIFANO_API_KEY}"},
    json={
        "input": "aifano://contract-bundle.pdf",
        "split_description": [
            {"title": "Master Service Agreement", "description": "The main service agreement"},
            {"title": "Statement of Work", "description": "Scope, deliverables, and timeline"},
            {"title": "Non-Disclosure Agreement", "description": "Confidentiality terms"}
        ]
    }
).json()

for section in split_result["result"]:
    print(f"Section: {section['category']} (pages {section['page_range']['start']}-{section['page_range']['end']})")

Tips

Use jobid:// to save credits

Parse once, then extract multiple times with different schemas using jobid://. This avoids re-parsing and saves 1 credit per page each time.

Be specific in your schema descriptions

Add description fields to guide extraction. For example, specify date formats, currency expectations, or what constitutes a “termination clause.”

Combine Split + Extract for bundled contracts

Split the document first to identify sections, then extract from each section individually for more accurate results.

Next Steps

Invoice Processing — Extract financial data from invoices
Multi-Document Pipelines — Build automated document workflows

​Overview

​What You’ll Build

​Step 1: Parse the Contract

​Step 2: Extract Key Contract Data

​Step 3: Example Output

​Step 4: Split Multi-Section Contracts

​Tips

​Next Steps