mirror of
https://github.com/Shubhamsaboo/awesome-llm-apps.git
synced 2026-04-29 14:48:44 -05:00
Working implementation of the project AI Legal Agent Team
This commit is contained in:
Madhu
58
ai_agent_tutorials/ai_legal_agent_team/README.md
Normal file
58
ai_agent_tutorials/ai_legal_agent_team/README.md
Normal file
@@ -0,0 +1,58 @@
|
||||
# AI Legal Agent Team
|
||||
|
||||
An intelligent legal document analysis Agent Team powered by GPT-4 and Qdrant vector database. The system uses a team of specialized AI agents to analyze legal documents, providing comprehensive insights, key points, and recommendations. We used phi-agent to create the agent team.
|
||||
|
||||
## Demo:
|
||||
|
||||
## Features
|
||||
|
||||
- **Specialized Agent Team**
|
||||
- Legal Researcher: Equipped with DuckDuckGo search tool to find and cite relevant legal cases and precedents. Provides detailed research summaries with sources and references specific sections from uploaded documents.
|
||||
|
||||
- Contract Analyst: Specializes in thorough contract review, identifying key terms, obligations, and potential issues. References specific clauses from documents for detailed analysis.
|
||||
|
||||
- Legal Strategist: Focuses on developing comprehensive legal strategies, providing actionable recommendations while considering both risks and opportunities.
|
||||
|
||||
- Team Lead: Coordinates analysis between team members, ensures comprehensive responses, properly sourced recommendations, and references to specific document parts. Acts as an Agent Team coordinator for all three agents.
|
||||
|
||||
- **Document Analysis Types**
|
||||
- Contract Review - Done by Contract Analyst
|
||||
- Legal Research - Done by Legal Researcher
|
||||
- Risk Assessment - Done by Legal Strategist, Contract Analyst
|
||||
- Compliance Check - Done by Legal Strategist, Legal Researcher, Contract Analyst
|
||||
- Custom Queries - Done by Agent Team - Legal Researcher, Legal Strategist, Contract Analyst
|
||||
|
||||
## How to Run
|
||||
|
||||
1. **Setup Environment**
|
||||
```bash
|
||||
# Clone the repository
|
||||
git clone https://github.com/Shubhamsaboo/awesome-llm-apps.git
|
||||
cd ai_legal_agent_team
|
||||
|
||||
# Install dependencies
|
||||
pip install -r requirements.txt
|
||||
```
|
||||
|
||||
2. **Configure API Keys**
|
||||
- Get OpenAI API key from [OpenAI Platform](https://platform.openai.com)
|
||||
- Get Qdrant API key and URL from [Qdrant Cloud](https://cloud.qdrant.io)
|
||||
|
||||
3. **Run the Application**
|
||||
```bash
|
||||
streamlit run main.py
|
||||
```
|
||||
4. **Use the Interface**
|
||||
- Enter API credentials
|
||||
- Upload a legal document (PDF)
|
||||
- Select analysis type
|
||||
- Add custom queries if needed
|
||||
- View analysis results
|
||||
|
||||
## Notes
|
||||
|
||||
- Supports PDF documents only
|
||||
- Uses GPT-4 for analysis
|
||||
- Uses text-embedding-3-small for embeddings
|
||||
- Requires stable internet connection
|
||||
- API usage costs apply
|
||||
335
ai_agent_tutorials/ai_legal_agent_team/main.py
Normal file
335
ai_agent_tutorials/ai_legal_agent_team/main.py
Normal file
@@ -0,0 +1,335 @@
|
||||
import streamlit as st
|
||||
from phi.agent import Agent
|
||||
from phi.knowledge.pdf import PDFKnowledgeBase, PDFReader
|
||||
from phi.vectordb.qdrant import Qdrant
|
||||
from phi.tools.duckduckgo import DuckDuckGo
|
||||
from phi.model.openai import OpenAIChat
|
||||
from phi.embedder.openai import OpenAIEmbedder
|
||||
import tempfile
|
||||
import os
|
||||
#initializing the session state variables
|
||||
def init_session_state():
|
||||
"""Initialize session state variables"""
|
||||
if 'openai_api_key' not in st.session_state:
|
||||
st.session_state.openai_api_key = None
|
||||
if 'qdrant_api_key' not in st.session_state:
|
||||
st.session_state.qdrant_api_key = None
|
||||
if 'qdrant_url' not in st.session_state:
|
||||
st.session_state.qdrant_url = None
|
||||
if 'vector_db' not in st.session_state:
|
||||
st.session_state.vector_db = None
|
||||
if 'legal_team' not in st.session_state:
|
||||
st.session_state.legal_team = None
|
||||
if 'knowledge_base' not in st.session_state:
|
||||
st.session_state.knowledge_base = None
|
||||
|
||||
def init_qdrant():
|
||||
"""Initialize Qdrant vector database"""
|
||||
if not st.session_state.qdrant_api_key:
|
||||
raise ValueError("Qdrant API key not provided")
|
||||
if not st.session_state.qdrant_url:
|
||||
raise ValueError("Qdrant URL not provided")
|
||||
|
||||
return Qdrant( #from the phidata Qdrant docs
|
||||
collection="legal_knowledge",
|
||||
url=st.session_state.qdrant_url,
|
||||
api_key=st.session_state.qdrant_api_key,
|
||||
https=True,
|
||||
timeout=None,
|
||||
distance="cosine"
|
||||
)
|
||||
|
||||
def process_document(uploaded_file, vector_db: Qdrant):
|
||||
"""Process document, create embeddings and store in Qdrant vector database"""
|
||||
if not st.session_state.openai_api_key:
|
||||
raise ValueError("OpenAI API key not provided")
|
||||
|
||||
os.environ['OPENAI_API_KEY'] = st.session_state.openai_api_key
|
||||
|
||||
with tempfile.TemporaryDirectory() as temp_dir:
|
||||
|
||||
temp_file_path = os.path.join(temp_dir, uploaded_file.name)
|
||||
with open(temp_file_path, "wb") as f:
|
||||
f.write(uploaded_file.getbuffer())
|
||||
|
||||
try:
|
||||
|
||||
embedder = OpenAIEmbedder(
|
||||
model="text-embedding-3-small",
|
||||
api_key=st.session_state.openai_api_key
|
||||
)
|
||||
|
||||
# Creating knowledge base with explicit Qdrant configuration
|
||||
knowledge_base = PDFKnowledgeBase(
|
||||
path=temp_dir,
|
||||
vector_db=vector_db,
|
||||
reader=PDFReader(chunk=True),
|
||||
embedder=embedder,
|
||||
recreate_vector_db=True
|
||||
)
|
||||
knowledge_base.load()
|
||||
return knowledge_base
|
||||
except Exception as e:
|
||||
raise Exception(f"Error processing document: {str(e)}")
|
||||
|
||||
def main():
|
||||
st.set_page_config(page_title="Legal Document Analyzer", layout="wide")
|
||||
init_session_state()
|
||||
|
||||
st.title("AI Legal Agent Team")
|
||||
|
||||
with st.sidebar:
|
||||
st.header("🔑 API Configuration")
|
||||
|
||||
openai_key = st.text_input(
|
||||
"OpenAI API Key",
|
||||
type="password",
|
||||
value=st.session_state.openai_api_key if st.session_state.openai_api_key else "",
|
||||
help="Enter your OpenAI API key"
|
||||
)
|
||||
if openai_key:
|
||||
st.session_state.openai_api_key = openai_key
|
||||
|
||||
qdrant_key = st.text_input(
|
||||
"Qdrant API Key",
|
||||
type="password",
|
||||
value=st.session_state.qdrant_api_key if st.session_state.qdrant_api_key else "",
|
||||
help="Enter your Qdrant API key"
|
||||
)
|
||||
if qdrant_key:
|
||||
st.session_state.qdrant_api_key = qdrant_key
|
||||
|
||||
qdrant_url = st.text_input(
|
||||
"Qdrant URL",
|
||||
value=st.session_state.qdrant_url if st.session_state.qdrant_url else "https://f499085c-b4bf-4bda-a9a5-227f62a9ca20.us-west-2-0.aws.cloud.qdrant.io:6333",
|
||||
help="Enter your Qdrant instance URL"
|
||||
)
|
||||
if qdrant_url:
|
||||
st.session_state.qdrant_url = qdrant_url
|
||||
|
||||
if all([st.session_state.qdrant_api_key, st.session_state.qdrant_url]):
|
||||
try:
|
||||
if not st.session_state.vector_db:
|
||||
st.session_state.vector_db = init_qdrant()
|
||||
st.success("Successfully connected to Qdrant!")
|
||||
except Exception as e:
|
||||
st.error(f"Failed to connect to Qdrant: {str(e)}")
|
||||
|
||||
st.divider()
|
||||
|
||||
if all([st.session_state.openai_api_key, st.session_state.vector_db]):
|
||||
st.header("📄 Document Upload")
|
||||
uploaded_file = st.file_uploader("Upload Legal Document", type=['pdf'])
|
||||
|
||||
if uploaded_file:
|
||||
with st.spinner("Processing document..."):
|
||||
try:
|
||||
knowledge_base = process_document(uploaded_file, st.session_state.vector_db)
|
||||
st.session_state.knowledge_base = knowledge_base
|
||||
|
||||
# Initialize agents
|
||||
legal_researcher = Agent(
|
||||
name="Legal Researcher",
|
||||
role="Legal research specialist",
|
||||
model=OpenAIChat(model="gpt-4"),
|
||||
tools=[DuckDuckGo()],
|
||||
knowledge=st.session_state.knowledge_base,
|
||||
search_knowledge=True,
|
||||
instructions=[
|
||||
"Find and cite relevant legal cases and precedents",
|
||||
"Provide detailed research summaries with sources",
|
||||
"Reference specific sections from the uploaded document",
|
||||
"Always search the knowledge base for relevant information"
|
||||
],
|
||||
show_tool_calls=True,
|
||||
markdown=True
|
||||
)
|
||||
|
||||
contract_analyst = Agent(
|
||||
name="Contract Analyst",
|
||||
role="Contract analysis specialist",
|
||||
model=OpenAIChat(model="gpt-4"),
|
||||
knowledge=knowledge_base,
|
||||
search_knowledge=True,
|
||||
instructions=[
|
||||
"Review contracts thoroughly",
|
||||
"Identify key terms and potential issues",
|
||||
"Reference specific clauses from the document"
|
||||
],
|
||||
markdown=True
|
||||
)
|
||||
|
||||
legal_strategist = Agent(
|
||||
name="Legal Strategist",
|
||||
role="Legal strategy specialist",
|
||||
model=OpenAIChat(model="gpt-4"),
|
||||
knowledge=knowledge_base,
|
||||
search_knowledge=True,
|
||||
instructions=[
|
||||
"Develop comprehensive legal strategies",
|
||||
"Provide actionable recommendations",
|
||||
"Consider both risks and opportunities"
|
||||
],
|
||||
markdown=True
|
||||
)
|
||||
|
||||
# Legal Agent Team
|
||||
st.session_state.legal_team = Agent(
|
||||
name="Legal Team Lead",
|
||||
role="Legal team coordinator",
|
||||
model=OpenAIChat(model="gpt-4"),
|
||||
team=[legal_researcher, contract_analyst, legal_strategist],
|
||||
knowledge=st.session_state.knowledge_base,
|
||||
search_knowledge=True,
|
||||
instructions=[
|
||||
"Coordinate analysis between team members",
|
||||
"Provide comprehensive responses",
|
||||
"Ensure all recommendations are properly sourced",
|
||||
"Reference specific parts of the uploaded document",
|
||||
"Always search the knowledge base before delegating tasks"
|
||||
],
|
||||
show_tool_calls=True,
|
||||
markdown=True
|
||||
)
|
||||
|
||||
st.success("✅ Document processed and team initialized!")
|
||||
|
||||
except Exception as e:
|
||||
st.error(f"Error processing document: {str(e)}")
|
||||
|
||||
st.divider()
|
||||
st.header("🔍 Analysis Options")
|
||||
analysis_type = st.selectbox(
|
||||
"Select Analysis Type",
|
||||
[
|
||||
"Contract Review",
|
||||
"Legal Research",
|
||||
"Risk Assessment",
|
||||
"Compliance Check",
|
||||
"Custom Query"
|
||||
]
|
||||
)
|
||||
else:
|
||||
st.warning("Please configure all API credentials to proceed")
|
||||
|
||||
# Main content area
|
||||
if not all([st.session_state.openai_api_key, st.session_state.vector_db]):
|
||||
st.info("👈 Please configure your API credentials in the sidebar to begin")
|
||||
elif not uploaded_file:
|
||||
st.info("👈 Please upload a legal document to begin analysis")
|
||||
elif st.session_state.legal_team:
|
||||
st.header("Document Analysis")
|
||||
|
||||
analysis_configs = {
|
||||
"Contract Review": {
|
||||
"query": "Review this contract and identify key terms, obligations, and potential issues.",
|
||||
"agents": ["Contract Analyst"],
|
||||
"description": "Detailed contract analysis focusing on terms and obligations"
|
||||
},
|
||||
"Legal Research": {
|
||||
"query": "Research relevant cases and precedents related to this document.",
|
||||
"agents": ["Legal Researcher"],
|
||||
"description": "Research on relevant legal cases and precedents"
|
||||
},
|
||||
"Risk Assessment": {
|
||||
"query": "Analyze potential legal risks and liabilities in this document.",
|
||||
"agents": ["Contract Analyst", "Legal Strategist"],
|
||||
"description": "Combined risk analysis and strategic assessment"
|
||||
},
|
||||
"Compliance Check": {
|
||||
"query": "Check this document for regulatory compliance issues.",
|
||||
"agents": ["Legal Researcher", "Contract Analyst", "Legal Strategist"],
|
||||
"description": "Comprehensive compliance analysis"
|
||||
},
|
||||
"Custom Query": {
|
||||
"query": None,
|
||||
"agents": ["Legal Researcher", "Contract Analyst", "Legal Strategist"],
|
||||
"description": "Custom analysis using all available agents"
|
||||
}
|
||||
}
|
||||
|
||||
st.info(f"📋 {analysis_configs[analysis_type]['description']}")
|
||||
st.write(f"🤖 Active Agents: {', '.join(analysis_configs[analysis_type]['agents'])}") #dictionary!!
|
||||
|
||||
user_query = st.text_area(
|
||||
"Enter your specific query:",
|
||||
help="Add any specific questions or points you want to analyze"
|
||||
)
|
||||
|
||||
if st.button("Analyze"):
|
||||
if user_query or analysis_type != "Custom Query":
|
||||
with st.spinner("Analyzing document..."):
|
||||
try:
|
||||
# Ensure OpenAI API key is set
|
||||
os.environ['OPENAI_API_KEY'] = st.session_state.openai_api_key
|
||||
|
||||
# Combine predefined and user queries
|
||||
if analysis_type != "Custom Query":
|
||||
combined_query = f"""
|
||||
Using the uploaded document as reference:
|
||||
|
||||
Primary Analysis Task: {analysis_configs[analysis_type]['query']}
|
||||
Additional User Query: {user_query if user_query else 'None'}
|
||||
|
||||
Focus Areas: {', '.join(analysis_configs[analysis_type]['agents'])}
|
||||
|
||||
Please search the knowledge base and provide specific references from the document.
|
||||
"""
|
||||
else:
|
||||
combined_query = user_query
|
||||
|
||||
response = st.session_state.legal_team.run(combined_query)
|
||||
|
||||
# Display results in tabs
|
||||
tabs = st.tabs(["Analysis", "Key Points", "Recommendations"])
|
||||
|
||||
with tabs[0]:
|
||||
st.markdown("### Detailed Analysis")
|
||||
if response.content:
|
||||
st.markdown(response.content)
|
||||
else:
|
||||
for message in response.messages:
|
||||
if message.role == 'assistant' and message.content:
|
||||
st.markdown(message.content)
|
||||
|
||||
with tabs[1]:
|
||||
st.markdown("### Key Points")
|
||||
key_points_response = st.session_state.legal_team.run(
|
||||
f"""Based on this previous analysis:
|
||||
{response.content}
|
||||
|
||||
Please summarize the key points in bullet points.
|
||||
Focus on insights from: {', '.join(analysis_configs[analysis_type]['agents'])}"""
|
||||
)
|
||||
if key_points_response.content:
|
||||
st.markdown(key_points_response.content)
|
||||
else:
|
||||
for message in key_points_response.messages:
|
||||
if message.role == 'assistant' and message.content:
|
||||
st.markdown(message.content)
|
||||
|
||||
with tabs[2]:
|
||||
st.markdown("### Recommendations")
|
||||
recommendations_response = st.session_state.legal_team.run(
|
||||
f"""Based on this previous analysis:
|
||||
{response.content}
|
||||
|
||||
What are your key recommendations based on the analysis, the best course of action?
|
||||
Provide specific recommendations from: {', '.join(analysis_configs[analysis_type]['agents'])}"""
|
||||
)
|
||||
if recommendations_response.content:
|
||||
st.markdown(recommendations_response.content)
|
||||
else:
|
||||
for message in recommendations_response.messages:
|
||||
if message.role == 'assistant' and message.content:
|
||||
st.markdown(message.content)
|
||||
|
||||
except Exception as e:
|
||||
st.error(f"Error during analysis: {str(e)}")
|
||||
else:
|
||||
st.warning("Please enter a query or select an analysis type")
|
||||
else:
|
||||
st.info("Please upload a legal document to begin analysis")
|
||||
|
||||
if __name__ == "__main__":
|
||||
main()
|
||||
5
ai_agent_tutorials/ai_legal_agent_team/requirements.txt
Normal file
5
ai_agent_tutorials/ai_legal_agent_team/requirements.txt
Normal file
@@ -0,0 +1,5 @@
|
||||
phidata==2.5.33
|
||||
streamlit==1.40.2
|
||||
qdrant-client==1.12.1
|
||||
openai
|
||||
|
||||
Reference in New Issue
Block a user