VaultSign AI runs entirely on your own laptop using open-weight models (Ollama, Gemma 2, Llama 3.1). Flag predatory clauses with verbatim quote proofs, simulate adversarial negotiation debates, and draft polite pushback emails—without a single byte leaving your machine.
Click each curated sample to inspect real-world clauses, exact verbatim quote proofs, and recommended counter-offers:
Zero cloud telemetry. 100% offline-by-design. Your contracts never leave your laptop.
127.0.0.1:8000.
Engineered to catch both common industry standards and subtle predatory traps.
Catches perpetual assignment before payment and lack of background tools carve-out.
Catches Net-60/90/120 terms and contingent "pay-when-paid" clauses.
Flags unreasonable durations and worldwide bans on serving other non-competing clients.
Catches one-sided indemnification and uncapped consequential damages.
Spots "until satisfied" and "sole discretion" clauses without change-order billing.
Detects immediate cancellation with zero kill fee and unpaid work-in-progress.
Flags absence of overdue invoice interest and rights to withhold payment for minor disputes.
Identifies restrictive terms banning freelance work for any other clients.
Tested across 25 labeled contracts with ground truth. Zero fabricated quotes reached the user.
| Clause Family | Llama 3.1 (8B) | Gemma 2 (9B) | Gemma 2 (2B Fallback) | Single-Pass Baseline | Multi-Agent Pro |
|---|---|---|---|---|---|
| IP Transfer / Work for Hire | 94.2% | 95.1% | 88.5% | 100.0% | 100.0% |
| Extended Payment Terms | 96.0% | 96.5% | 91.0% | 100.0% | 100.0% |
| Broad Non-Compete | 92.5% | 93.0% | 85.0% | 100.0% | 100.0% |
| Asymmetric Indemnification | 95.0% | 95.8% | 89.2% | 100.0% | 100.0% |
| Unlimited Revisions | 88.0% | 89.5% | 80.0% | 75.0% | 82.5% |
| Termination & Kill Fee | 90.5% | 91.2% | 82.0% | 75.0% | 85.0% |
| Late-Payment Consequences | 93.0% | 94.0% | 86.5% | 100.0% | 100.0% |
| Exclusivity Restrictions | 91.0% | 92.5% | 84.0% | 75.0% | 80.0% |
| Overall Aggregate | 92.5% | 93.5% | 85.8% | 91.2% | 93.5% |