# Project export: OMNI-ANALYST
This document was generated by HackStack to give an AI agent context about a hackathon project. Sections are labeled with their provenance; content marked as truncated was cut to keep this document small.
## Project metadata
- Hackathon: TreeHacks 2026
- Tagline: Omni-Analyst: Autonomous Medical-Legal Agent. Automating FDA compliance for clinical trials via Claude Agent SDK & Modal. Reducing months of manual audit to seconds. Scalable, Secure, Agentic.
- Devpost: https://devpost.com/software/omni-analyst
- GitHub: https://github.com/RobertUmniy/med-ai-for-lungs
- Team: 0 GitHub contributor(s) —
## Devpost submission (written by the team)
### Overview
OMNI-ANALYST: Autonomous Regulatory Intelligence
### Inspiration
Clinical trials are delayed by months due to manual regulatory review. Compliance teams are drowning in 500-page protocols while trying to keep up with evolving FDA and international standards. We asked: What if regulatory intelligence could be autonomous? Omni-Analyst was born from the gap between life-saving innovation and slow compliance infrastructure. We built an AI agent that doesn't just chat—it reasons over medical and legal documents at enterprise scale.
### What it does
Omni-Analyst is an autonomous medical-legal AI agent that: Parses 500+ page clinical trial protocols in seconds. Cross-references them with complex regulatory frameworks (FDA, EMA, global standards). Detects inconsistencies, compliance risks, and missing disclosures. Generates structured, audit-ready compliance reports.
### How we built it
We designed Omni-Analyst as a multi-agent reasoning system to ensure professional-grade accuracy. The Tech Stack: LLM-powered reasoning core for specialized legal-medical analysis. Retrieval Layer using vector embeddings over massive regulatory databases. Structured Reasoning Pipeline to eliminate hallucinations and maintain context. Autonomous Modules for risk flagging and report synthesis.
### Challenges we ran into
The biggest hurdle was preventing hallucinations in high-stakes regulatory language. We solved this by: Implementing retrieval-constrained generation. Using structured prompts with legal reasoning chains. Building multi-step validation gates before any report is finalized.
### Accomplishments we're proud of
Built a functioning multi-agent compliance prototype in under 48 hours. Achieved structured, audit-style output instead of generic conversational responses. Designed a scalable architecture suitable for enterprise-level biotech deployment. The system behaves like a regulatory analyst, not just a chatbot.
### What we learned
Agent orchestration matters more than model size. Designing structured reasoning flows dramatically improves reliability in high-risk domains like healthcare. AI is most powerful when it augments professional workflows rather than replacing human judgment.
### What's next
Global Expansion: Adding support for additional international jurisdictions. Real-time Intelligence: Integrating live regulatory update feeds. Security: Building secure, air-gapped infrastructure for sensitive data. Partnerships: Pilot programs with biotech firms and legal teams.
## README (from the GitHub repository)
No README available.
## Detected evidence (automated analysis)
Indexed codebase: 1 recognized source files, 3 KB.
- HTML (language) — detected in the code
## Codebase structure (from repository index)
### Files (1 of 1)
```
MEDAIFORL.html
```
### Dependencies
No dependency index available.
### Recent commits (newest first)
- Add files via upload
## Key source files (fetched from GitHub, selected and truncated for size)
### MEDAIFORL.html
```html
MED AI FOR LUNGS
MED AI FOR LUNGS
Искусственный интеллект, который анализирует снимки лёгких и выявляет заболевания за секунды.
Используем самые современные модели ИИ и компьютерного зрения.
⚡ Мгновенный результат
Диагностика за секунды — от простуды до онкологии.
🌍 Доступность
Помогаем людям по всему миру получать раннюю диагностику.
Как это работает?
Загружаете медицинский снимок — система анализирует его с помощью нейросетей и показывает вероятный диагноз.
MED AI FOR LUNGS способен различать рак, пневмонию, туберкулёз, COVID-19, грипп и другие болезни лёгких.