Domain

LegalForge AI

LegalForge automates legal document analysis in four stages: 1. **Upload** — ingest PDF legal documents into a Qdrant vector knowledge base 2. **Research** — Legal Researcher agent finds cases, precedents, and citations via DuckDuckGo + knowledge base 3. **Analyze** — Contract Analyst and Legal Strategist review terms, risks, and strategy 4. **Report** — Team Lead synthesizes analysis, key points, and recommendations

Python Streamlit

← Back to marketplace

LegalForge AI

Multi-agent legal document analysis by Sharad Khare.

LegalForge AI is a coordinated legal agent team that indexes PDF documents in Qdrant, searches the web with DuckDuckGo, and delivers contract review, legal research, risk assessment, and compliance analysis through Streamlit.

Built and maintained by Sharad Khare — AI strategist, full-stack developer, and creator of practical agent workflows.

What this project does

LegalForge automates legal document analysis in four stages:

1. Upload — ingest PDF legal documents into a Qdrant vector knowledge base

2. Research — Legal Researcher agent finds cases, precedents, and citations via DuckDuckGo + knowledge base

3. Analyze — Contract Analyst and Legal Strategist review terms, risks, and strategy

4. Report — Team Lead synthesizes analysis, key points, and recommendations

Why Sharad Khare built LegalForge AI

Legal review is time-intensive and requires cross-disciplinary expertise — research, contract parsing, and strategic advice. LegalForge demonstrates how a coordinated Agno agent team with vector search compresses that workflow into a single Streamlit experience.

How it works

PDF Upload
   │
   ▼
Qdrant Vector Index (OpenAI embeddings)
   │
   ▼
Legal Agent Team
   ├── Legal Researcher (DuckDuckGo + knowledge base)
   ├── Contract Analyst (clause and term review)
   └── Legal Strategist (risk and recommendations)
   │
   ▼
Team Lead → Analysis, Key Points, Recommendations

Agent roles

| Agent | Tools | Output |

|-------|-------|--------|

| Legal Researcher | DuckDuckGo, Qdrant KB | Case law and precedent research |

| Contract Analyst | Qdrant KB | Clause review and term analysis |

| Legal Strategist | Qdrant KB | Risk assessment and strategy |

| Team Lead | Coordinates all agents | Comprehensive legal report |

Analysis types

Contract Review
Legal Research
Risk Assessment
Compliance Check
Custom Query

Features

PDF upload with Qdrant vector indexing
Three specialized legal agents plus Team Lead coordinator
DuckDuckGo web research combined with document knowledge base
Five analysis modes from contract review to custom queries
Tabbed output: detailed analysis, key points, recommendations

Quick start

cd legalforge-ai
pip install -r requirements.txt
streamlit run app.py

Required API keys

OpenAI API key — agent reasoning and embeddings
Qdrant API key — vector database for document search
Qdrant URL — your Qdrant Cloud or self-hosted instance URL

Project structure

legalforge-ai/
├── app.py
├── legalforge/
│   ├── config.py
│   ├── documents.py
│   ├── agents.py
│   └── analysis.py
├── requirements.txt
├── pyproject.toml
└── README.md

Topics

Sharad KhareLegalForge AIlegal document analysis AIcontract review agentQdrant legal knowledge basemulti-agent legal teamlegal research automationcompliance check AIrisk assessment agentStreamlit legal analyzerAgno legal agents.

Who this is for

Legal professionals reviewing contracts and agreements
Compliance teams checking regulatory alignment
Startup founders evaluating vendor and partnership contracts
Developers learning multi-agent + vector search patterns for legal tech

Use cases

Upload a contract PDF and run automated clause review
Research relevant case law alongside uploaded documents
Assess legal risks and liabilities in agreements
Run compliance checks against regulatory requirements
Ask custom legal questions with full team coordination

Author

Website: sharadkhare.in

License

Loading file tree…

Select a file to preview its source.

bash

cd legalforge-ai
pip install -r requirements.txt
streamlit run app.py