$ cat projects/lexa-legal-ai.md
Lexa — Legal Research AI That Abstains
A grounded legal AI that says "I don't know" instead of hallucinating.
A legal research assistant built on a 3-agent pipeline — a Researcher retrieves sources, a Drafter writes with inline citations, and an independent Verifier checks every claim. If the evidence is weak, Lexa abstains rather than guessing. The whole pipeline streams live to the UI.
// problem
LLMs hallucinate citations — unacceptable in legal research, where one fabricated case can cost a career.
// impact
A deterministic citation guard enforces that every claim maps to a retrieved source; a score-floor abstention gate refuses low-evidence answers; a bounded repair loop fixes draft issues before anything reaches the user. Ships offline out-of-the-box (mock model + local embeddings + a bundled US Constitutional corpus), with optional live connectors to Federal Register, CourtListener, and GovInfo. pytest suite plus live-source E2E tests on GitHub Actions CI.
// tech-stack