ContractGuard

AI-powered contract review agent — never sign a bad contract again.

Upload any contract → get red flags, unfair terms, and plain-English explanations in seconds.

English | 中文

Why ContractGuard?

Every year, millions of people sign contracts they don't fully understand — apartment leases with hidden penalties, employment agreements with overly broad non-competes, NDAs that silently strip away your rights. Hiring a lawyer costs $300-500/hour. Most people just sign and hope for the best.

ContractGuard changes that. It's an open-source AI agent that reads every clause of your contract, flags problems in plain language, and tells you exactly what to negotiate — all in under 30 seconds.

What makes it different from ChatGPT?

• Structured analysis, not a wall of text — you get categorized red flags, warnings, protections, and a fairness score
• Actionable suggestions for every issue found — not just "this is bad" but "change it to this"
• Consistent output format via Pydantic models — easy to integrate into other tools
• CLI-first design — one command, beautiful terminal output, no browser needed
• Works with any LLM — OpenRouter, OpenAI, Ollama (fully local/private)

Demo

contractguard scan my-lease.pdf

✔ Parsed my-lease.pdf (4,521 characters)

⬤ RED FLAGS (5 found)
==================================================

  1. Non-refundable security deposit
     Clause: Section 3
     "The security deposit is non-refundable and shall
      be retained by Landlord upon termination"
     Most states require deposits to be refundable.
     This clause is likely illegal in California.
     Suggestion: Remove "non-refundable" language.

  2. Unlimited landlord access without notice
     Clause: Section 5
     "Landlord shall have the right to enter the Property
      at any time, with or without notice"
     California law requires 24-hour written notice.
     Suggestion: Add "with 24 hours written notice"

  3. Tenant pays for structural repairs
     Clause: Section 4
     "Tenant is responsible for all maintenance and repairs
      including plumbing, electrical systems, and structural elements"
     Structural repairs are typically the landlord's responsibility.
     Suggestion: Limit tenant responsibility to minor maintenance only.

  4. One-sided termination clause
     Clause: Section 10
     "Landlord may terminate this Lease at any time with
      thirty (30) days' written notice for any reason"
     Tenant has no equivalent right. This creates an imbalance.
     Suggestion: Add mutual termination rights or require cause.

  5. Tenant pays landlord's attorney fees regardless of outcome
     Clause: Section 12
     "Tenant shall be responsible for all attorney's fees
      incurred by Landlord, regardless of the outcome"
     One-sided fee-shifting discourages tenants from asserting rights.
     Suggestion: Change to mutual fee-shifting (loser pays).

⚠ WARNINGS (3 found)
==================================================

  1. 90-day notice period for non-renewal
     Clause: Section 1
     Suggestion: Negotiate down to 30-60 days.

  2. 15% rent increase cap on renewal
     Clause: Section 2
     Suggestion: Negotiate a lower cap or tie to CPI.

  3. No subletting without landlord consent (may withhold for any reason)
     Clause: Section 9
     Suggestion: Add "consent shall not be unreasonably withheld."

✔ PROTECTIONS (2 found)
==================================================
  ✔ Written notice required for termination (Section 1)
  ✔ Lease modification requires written agreement from both parties (Section 13)

❓ MISSING PROTECTIONS (4)
  ✗ No habitability guarantee
  ✗ No grace period for lease violations before termination
  ✗ No limit on late fees
  ✗ No provision for return of security deposit after move-out

FAIRNESS SCORE: D (28/100)
  5 red flags  3 warnings  2 protections  4 missing

Quick Start

1. Install

pip install contractguard

2. Set up your API key

ContractGuard works with any OpenAI-compatible API. Pick one:

Option A: OpenRouter (recommended) — access to Claude, GPT-4, DeepSeek, Gemini, and 100+ models through a single API key:

export OPENROUTER_API_KEY=sk-or-...

Option B: OpenAI directly:

export OPENAI_API_KEY=sk-...
export OPENAI_BASE_URL=https://api.openai.com/v1

Option C: Local models via Ollama — your contract data never leaves your machine:

export OPENAI_BASE_URL=http://localhost:11434/v1
export OPENAI_API_KEY=ollama

3. Scan a contract

contractguard scan my-contract.pdf

That's it. Three steps, under 60 seconds.

Usage

CLI Commands

# Basic scan — PDF, DOCX, or TXT
contractguard scan lease.pdf
contractguard scan employment-agreement.docx
contractguard scan nda.txt

# Choose a specific model
contractguard scan contract.pdf --model openai/gpt-4o
contractguard scan contract.pdf --model anthropic/claude-sonnet-4
contractguard scan contract.pdf --model google/gemini-2.5-pro
contractguard scan contract.pdf --model llama3.1    # local via Ollama

# Export the report as a markdown file
contractguard scan contract.pdf --output report.md

# Get structured JSON output (for integrations or scripting)
contractguard scan contract.pdf --json

# Pass API key directly (instead of env variable)
contractguard scan contract.pdf --api-key sk-or-...

Python API

Use ContractGuard as a library in your own applications:

from contractguard.analyzer import analyze_contract
from contractguard.parser import extract_text

# Step 1: Extract text from any supported document
text = extract_text("my-lease.pdf")

# Step 2: Run the AI analysis
result = analyze_contract(text)

# Step 3: Use the structured results
print(f"Contract Type: {result.contract_type.value}")
print(f"Fairness Score: {result.fairness_grade} ({result.fairness_score}/100)")
print(f"Parties: {', '.join(result.parties)}")

print(f"\n{len(result.red_flags)} Red Flags:")
for flag in result.red_flags:
    print(f"  - {flag.title} (Clause: {flag.clause})")
    print(f"    Issue: {flag.explanation}")
    print(f"    Fix: {flag.suggestion}")

print(f"\n{len(result.warnings)} Warnings:")
for warning in result.warnings:
    print(f"  - {warning.title}: {warning.explanation}")

print(f"\n{len(result.good_clauses)} Protections:")
for protection in result.good_clauses:
    print(f"  + {protection.title}: {protection.explanation}")

print(f"\n{len(result.missing_protections)} Missing Protections:")
for missing in result.missing_protections:
    print(f"  ? {missing}")

# Export to markdown
from contractguard.report import generate_markdown_report
md = generate_markdown_report(result)
with open("report.md", "w") as f:
    f.write(md)

JSON Output Schema

When using --json, ContractGuard outputs a structured JSON object you can pipe to other tools:

{
  "contract_type": "lease",
  "summary": "A 12-month residential lease with several one-sided clauses...",
  "parties": ["Apex Property Management LLC", "Tenant"],
  "key_terms": ["Duration: 12 months", "Rent: $3,200/month", "Deposit: $6,400"],
  "red_flags": [
    {
      "title": "Non-refundable security deposit",
      "severity": "red",
      "clause": "Section 3",
      "quote": "The security deposit is non-refundable...",
      "explanation": "Most states require deposits to be refundable...",
      "suggestion": "Remove non-refundable language."
    }
  ],
  "warnings": [...],
  "good_clauses": [...],
  "missing_protections": [...],
  "fairness_score": 28,
  "fairness_grade": "D"
}

Supported File Formats

| Format | Extension | Notes | |--------|-----------|-------| | PDF | .pdf | Text-based PDFs. Scanned/image-based PDFs require OCR (coming soon). | | Word | .docx | Microsoft Word documents | | Plain Text | .txt | Plain text files | | Markdown | .md | Markdown files | | Rich Text | .rtf | Rich Text Format files |

Supported Contract Types

ContractGuard automatically detects the contract type and tailors its analysis accordingly. Each type has specific red flags and industry-standard protections it checks for:

| Contract Type | What ContractGuard Checks | |---|---| | Residential Leases | Rent increases, deposit refundability, maintenance obligations, landlord access rights, early termination penalties, habitability guarantees | | NDAs / Confidentiality | Scope of "confidential information" (too broad?), duration, non-solicitation, non-compete, carve-outs for prior knowledge, return/destruction of materials | | Employment Contracts | Non-compete scope & duration, IP assignment (does employer own your side projects?), termination notice period, severance, at-will vs. for-cause, benefits | | Freelance / Contractor | Payment terms & schedule, kill fees, IP ownership, indemnification, scope creep protections, late payment penalties | | SaaS Terms of Service | Data ownership & portability, auto-renewal & cancellation, SLA guarantees, limitation of liability, unilateral modification rights | | Loan Agreements | Interest rate (fixed vs. variable), prepayment penalties, default triggers, personal guarantee scope, collateral requirements | | Purchase Agreements | Warranty terms, return/refund policy, liability limits, dispute resolution (arbitration vs. court), force majeure |

How It Works

1. Parse — Extracts text from your document (PDF, DOCX, TXT). For PDFs, uses pdfplumber to handle complex layouts. For DOCX, uses python-docx to read all paragraph