Extract structured data from contracts, legal agreements, NDAs, court filings, leases, and compliance documents. Then redact privileged or sensitive information before sharing with opposing counsel, auditors, or external parties.
This skill instructs the agent to POST documents to https://api.deepread.tech and poll for results. No system files are modified.
What You Get Back
Submit a contract and get structured JSON. Extracted fields come back as a list under extraction.fields[] (each field has key, value, needs_review, and location.page):
"Provider indemnifies Client against third-party IP claims"
,
"page"
:
5
}
,
{
"type"
:
"Limitation of Liability"
,
"summary"
:
"Capped at 12 months of fees paid"
,
"page"
:
5
}
,
{
"type"
:
"Termination"
,
"summary"
:
"Either party may terminate with 30 days written notice"
,
"page"
:
6
}
,
{
"type"
:
"Non-Compete"
,
"summary"
:
"12-month non-compete within same industry vertical"
,
"page"
:
7
}
]
,
"needs_review"
:
false
,
"location"
:
{
"page"
:
5
}
}
,
{
"key"
:
"signatures"
,
"value"
:
[
{
"name"
:
"John Smith"
,
"title"
:
"CEO, Acme Corp"
,
"date"
:
"2026-01-10"
}
,
{
"name"
:
"Jane Doe"
,
"title"
:
"General Counsel, GlobalCo Inc"
,
"date"
:
"2026-01-12"
}
]
,
"needs_review"
:
false
,
"location"
:
{
"page"
:
8
}
}
]
}
}
Fields with needs_review: true need human review (check review_reason). Everything else is high-confidence.
Setup
Get Your API Key
open "https://www.deepread.tech/dashboard/?utm_source=clawhub"
Save it:
export DEEPREAD_API_KEY="sk_live_your_key_here"
Contract Schema
Pre-built schema for contracts and legal agreements:
{"type":"object","properties":{"document_type":{"type":"string","description":"Type of legal document (MSA, NDA, lease, employment agreement, etc.)"},"parties":{"type":"array","items":{"type":"object","properties":{"name":{"type":"string","description":"Full legal name of the party"},"role":{"type":"string","description":"Role in the agreement (e.g., Licensor, Licensee, Landlord, Tenant)"},"address":{"type":"string","description":"Address of the party"}}},"description":"All parties to the agreement"},"effective_date":{"type":"string","description":"Date the agreement takes effect (YYYY-MM-DD)"},"termination_date":{"type":"string","description":"End date or expiration of the agreement (YYYY-MM-DD)"},"governing_law":{"type":"string","description":"Jurisdiction governing the agreement"},"contract_value":{"type":"number","description":"Total contract value or consideration amount"},"payment_terms":{"type":"string","description":"Payment schedule and terms"},"key_clauses":{"type":"array","items":{"type":"object","properties":{"type":{"type":"string","description":"Clause type (Indemnification, Limitation of Liability, Termination, Non-Compete, Confidentiality, IP Ownership, etc.)"},"summary":{"type":"string","description":"Brief summary of the clause terms"},"page":{"type":"number","description":"Page number where the clause appears"}}},"description":"Key clauses and their summaries"},"renewal_terms":{"type":"string","description":"Auto-renewal terms if applicable"},"notice_period":{"type":"string","description":"Required notice period for termination"},"signatures":{"type":"array","items":{"type":"object","properties":{"name":{"type":"string","description":"Signatory name"},"title":{"type":"string","description":"Title and organization"},"date":{"type":"string","description":"Date signed (YYYY-MM-DD)"}}},"description":"Signatories and execution dates"}}}
Specify clause types — Listing expected clause types (Indemnification, Limitation of Liability, etc.) in the schema description improves extraction
Use page numbers — Adding "page number where the clause appears" helps locate extracted terms in the original
Create blueprints for recurring document types — If you process the same type of contract repeatedly (e.g., NDAs, MSAs), train a blueprint at deepread.tech/dashboard/optimizer for 20-30% accuracy improvement
Always redact before external sharing — Use PII redaction before sending documents to opposing counsel, third-party reviewers, or LLMs
BYOK — Zero Processing Costs
Connect your own OpenAI, Google, or OpenRouter key via the dashboard. All document processing routes through your provider — zero DeepRead LLM costs, page quota skipped.
