Overview

A Chrome extension that uses AI agents to help users negotiate the best deals.

Connection test site

Table of Contents

generated with DocToc

• Table of Contents
• Overview
• Key Features
• Technologies Used
• Project Structure
• Setup
• Usage
• Development
• Package Management
• Customizing AI Agents
• Modifying Prompts
• Contributing
• Troubleshooting

Overview

A Chrome extension that uses AI agents to help the user negotiate the best deals:

1. Context & Goal Setting:

   • From the chrome extention, ask users about their goals of the deal.
   • Use Pydantic to define the schema via models

2. Negotiation Starts:

   • Employ Llama 3.1 (running on Together AI) and call /browse API endpoint via FastAPI
   • Record the log on the negotiation.log to track the conversation

3. User Interaction:

   • Presents the seller's response and ask for a counter response.
   • Allows the buyer to choose whether to accept or renegotiate.

4. Response Generation:

   • Collect user input.
   • Passes the information back to Llama 3.1 on Together AI.
   • Generates response to the end user.

Key Features

• Collect user inputs via Chrome extention
• Use Pydantic to build a schema and validate the inputs
• AI-driven context analysis & negotiation converation using Llama 3.1 on Together AI
• Call API via FastAPI and render response on the user-friendly interface
• Automated generation of response drafts using Llama 3.1

Technologies Used

[data-doc-management]

• Upstage: Converts natural language queries into SQL Doc
• Chroma DB: Vector database for storing and querying standard contract clauses
• SQLite: Database for storing application data

[ai-model-curation]

• Together AI: Hosting Llama 3.1 for text processing, clause segmentation, and response generation

[deployment-framework]

• Python: Primary programming language
• Poetry: Backend build system and package installer/resolver doc
• Pydantic: Data validation library for Python
• Fast API: Web framework for building APIs based on standard Python type hints
• Replit: Deploy Fast API on Replit server (See the .replit and replit.nix for its configuration.)
• Koyeb: Deploy the backend repo and access in the production env

Project Structure

.
└── frontend - react app
|
└── src
   └── __init__.py
   └── api.py = connect with database
   └── main.py = build fastAPI connection
└── Prompts
   └── Browse.py
   └── Negotiate.py
   └── Validate.py
└── tests
   └── __init__.py
   └── test_endpoint.py: test connection to the browser endpoint
   └── test_negotiation.py: test and log the negotiation process and outcome
├── pyproject.toml
└── poetry.lock

< Koyeb deployment>
└── requirements.txt
└── Procfile
└── runtime.txt

Setup

1. Install the poetry package manager:

   curl -sSL https://install.python-poetry.org | python -
   poetry --version
   

2. Install dependencies on pyproject.toml:

   poetry install
   

3. Set up environment variables: Create a .env file in the project root and add the following:

   TOGETHER_API_KEY=your_together_api_key
   DATABASE_URL=your_database_url
   API_ROUTE=your_public_api_route
   

Usage

1. Test the negotiation process and outcome on terminal:

   poetry run neg
   

   The outcome will be recorded in the negotiation.log file.

2. In a separate terminal, test the connection to the client interface:

   poetry run client
   

   The results will be printed on your terminal. When you update the main.py, run this command to see if the update is working properly.

   The backend will be available at http://localhost:8000.

3. Deploy the API via Koyab

   koyeb deploy . <amazing_app_name_of_your_choice>/src --instance-type=free
   

   *See the doc for the Koyeb CLI installation and more.

   Now you can access the API endpoint from the public URL ...koyab.app.

4. Run the frontend. (Open another terminal and set the root to frontend. For now, frontend contains test connection to the public API)

   npm i
   npm start 
   

   The frontend will be available at http://localhost:3000.

Development

Package Management with poetry

• Add a package: poetry add <package>
• Remove a package: poetry remove <package>
• Run a command in the virtual environment: add a script to pyproject.toml, then poetry run <command>
• To reinstall all the package: after deleting poetry.lock file, run poetry install

poetry cli

Customizing AI Agents

To modify or add new AI agents, edit the test_negotiation.py file to test on the terminal, and/or edit the main.py file to apply the modification to the API.

Modifying Prompts

Necessary prompts are stored in the Prompts folder. Add or modify the prompt accordingly.

(To avoid errors, we recommend adding a new prompt file to the Prompt folder.)

Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/your-amazing-feature)
3. Commit your changes (git commit -m 'Add your-amazing-feature')
4. Push to the branch (git push origin feature/your-amazing-feature)
5. Open a pull request

Troubleshooting

Common issues and solutions:

• API key errors: Ensure all API keys in the .env file are correct and up to date.
• Database connection issues: Check if the Chroma DB is properly initialized and accessible.
• Memory errors: If processing large contracts, you may need to increase the available memory for the Python process.
• Issues related to the AI agents: Check the negotiation.log file for detailed error messages and stack traces.