jeduclo/LexCam---Assistant-P-nal-Camerounais
Domain:
natural language processingRecord type:
software
Creator:
jed
Host:
LexCam is an advanced AI-powered legal assistant designed to help users navigate the Cameroon Penal Code (2016). It uses Retrieval-Augmented Generation (RAG) to provide accurate, citation-backed answers and simulates a legal tribunal for case analysis.
LexCam 🇨🇲 - Assistant Pénal Camerounais
LexCam is an advanced AI-powered legal assistant designed to help users navigate the Cameroon Penal Code. It uses Retrieval-Augmented Generation (RAG) to provide accurate, citation-backed answers and simulates a legal tribunal for case analysis.
Features
AI Chatbot: Ask questions in plain language and get answers grounded in the Penal Code.
Multi-Agent System:
Investigator Agent: Extracts legal keywords from user scenarios.
Judge Agent: Analyzes facts using the IRAC method (Issue, Rule, Analysis, Conclusion).
Defense Agent: Proposes mitigating circumstances and legal excuses.
Prosecution Agent: Argues for strict application of the law and aggravating circumstances.
Persistence: Automatically indexes and saves PDF documents to disk for instant loading.
Transparency: Strict hallucination controls (Temperature 0) and citation requirements.
Tech Stack
Python 3.9+
Streamlit: Web Interface
LangChain: Agent Orchestration
Google Gemini (1.5 Flash): LLM Engine
FAISS: Vector Database (Local Storage)
PyPDF: Document Processing
📦 Installation
1. Clone the Repository
git clone github.com
cd lexcam-cameroun
2. Set up Virtual Environment
Windows:
python -m venv venv
venv\Scripts\activate
3. Install Dependencies
pip install -r requirements.txt
4. Configure API Key
Create a .streamlit/secrets.toml file in the root directory:
# .streamlit/secrets.toml
GOOGLE_API_KEY = "your_google_api_key_here"
(Get your key from Google AI Studio)
5. Add Data
Ensure your PDF documents (e.g., Cameroun-Code-2016-penal1.pdf) are inside the data/ folder.
Usage
Run the application locally:
streamlit run app.py
The app will open in your browser at localhost.
First Run: The app will take a few minutes to process the PDFs and create the faiss_index_store.
Subsequent Runs: It will load instantly from the saved index.
⚖️ Disclaimer
This tool is an AI experiment and does not constit …