Why I Built VaultDocs AI: Local-First Contract Risk Screening with Open Models
Описание и применение
VaultDocs AI – локальное решение для проверки рисков в контрактах, использующее открытые модели LLM и обеспечивающее полную конфиденциальность данных.
Задача
Freelancers and small agencies need to review contracts for risky clauses but cannot afford legal review and cannot upload confidential documents to commercial cloud LLMs.
Результат
VaultDocs AI provides a privacy‑guaranteed, zero‑cloud contract risk screening workflow that runs locally, reduces cost to $0 per query, and delivers risk scores and audit reports without exposing confidential data.
Рабочий процесс
- Parse PDF contracts into page‑bound chunks using PyMuPDF.
- Rank clauses with an in‑memory BM25 + legal keyword hybrid ranker and FAISS vector index to retrieve risk‑prone sections in under 5 ms.
- Screen retrieved sections locally with open‑weight LLMs (llama3.2:1b, gemma:2b, qwen2.5:3b) via Ollama.
- Validate model citations against extracted PDF text to eliminate hallucinations.
- Render a professional 3‑column legal review workstation with customizable risk scoring (0–100) and downloadable Markdown audit reports.
Доказательства
VaultDocs AI is a local-first, privacy-guaranteed contract risk screening assistant tailored for freelancers, contractors, and boutique agency owners.
No contract text, vector embeddings, or prompts ever leave local memory.
Parses PDF contracts into page-boundary chunks.
Uses a fast in-memory hybrid ranker (BM25 + legal keyword boost) and dense FAISS vector index to retrieve risk-prone sections in under 5 milliseconds.
Performs local open-weight LLM screening using llama3.2:1b, gemma:2b, or qwen2.5:3b via Ollama.
Validates model citations directly against extracted PDF text to eliminate hallucinations.
Renders a professional, interactive 3-column legal review workstation with customizable risk scoring (0–100) and downloadable Markdown audit reports.
- Дата записи: 2026-10-15T17:41:01.289Z