Sahayak — An AI Legal Assistant That Says "I Don't Know" When It Should

A developer built Sahayak, a RAG-based legal assistant that answers tenancy, consumer-rights and contract questions and summarizes uploaded PDFs in plain language, grounding every response in retrieved context and explicitly refusing to answer when no matching context exists. The system pairs a thin FastAPI retrieval-plus-completion hot path with Groq's openai/gpt-oss-120b model, ChromaDB and sentence-transformers for retrieval, PyMuPDF and pytesseract for parsing, and Groq Whisper for voice input. The developer highlighted guardrails including an untrusted-context system prompt designed to resist prompt injection, python-magic byte-level upload validation, slowapi rate limiting, security headers, and five CI workflows covering lint, pytest, bandit, gitleaks, pip-audit and axe-core.

Legal help is expensive, jargon-heavy, and often out of reach for people who need it most. For PromptWars: Virtual Exclusive Edition — a hackathon by Hack2Skill — I built Sahayak , a RAG-based legal assistant that answers questions and summarizes documents in plain language, grounded in real context instead of confident guesses. 🔗 Live: https://legal-ai-eight-liart.vercel.app/ https://legal-ai-eight-liart.vercel.app/ 🔗 Code: https://github.com/Divyansh0208/Legal-AI https://github.com/Divyansh0208/Legal-AI MIT What it does - Ask — a tenancy, consumer-rights, or contract question. Answers are retrieved from indexed context, not hallucinated. No matching context? It says so, instead of guessing. - Upload — a PDF scanned or native → plain-language summary: what kind of document it is, what it obligates you to, what's worth double-checking. - Voice — ask by speaking, transcribed via Whisper. Every response is grounded in retrieved context and explicit about what it can't confirm — no confident guessing when the docs don't cover something. Architecture The RAG hot path is intentionally thin — no framework ceremony for a single retrieval + one chat completion call: The part I'm actually proud of: the guardrails Anyone can wire an LLM to a text box. The parts that took real effort: - Untrusted-context system prompt — uploaded documents and retrieved chunks are treated as data , never instructions. Prompt-injection resistant by design. - Real upload validation — python-magic sniffs actual file bytes; a relabeled .exe doesn't sneak through as .pdf . - Rate limiting — slowapi on query, upload, and voice endpoints. - Security headers — CSP, X-Frame-Options, HSTS in production. - 5 CI workflows on every push — lint, pytest, bandit SAST , gitleaks secrets , pip-audit deps , axe-core a11y — auto-deploy to Render/Vercel on merge to main . None of that shows up in a demo GIF. All of it is why the demo GIF is trustworthy. Stack | Layer | Tech | | Backend | FastAPI | | LLM | Groq openai/gpt-oss-120b | | Retrieval | ChromaDB + sentence-transformers all-MiniLM-L6-v2 | | Parsing | PyMuPDF, pytesseract | | Speech | Groq Whisper | | Frontend | React + Vite, Tailwind | | DB | SQLite | | Deploy | Render + Vercel | What's next - Multi-language support deferred - Persisted chat sessions currently in-memory - More jurisdiction-specific templates If you've hit a real legal-access gap, or want to poke holes in the grounding/refusal behavior — feedback welcome. MIT-licensed, issues/PRs open.