ai-bootstrap 2.2.0

AI-powered project scaffolding with requirement analysis

AI Bootstrap 🚀

Your AI-Powered Project Scaffolding CLI

AI Bootstrap is a professional CLI tool that uses AI to plan, scaffold, and generate production-ready AI/ML project blueprints. Go from a simple idea to a fully structured, modern Python application in seconds.

---

AI Bootstrap streamlines the initial, often tedious, phases of AI project development. Whether you prefer an AI-driven chat to plan your project or a guided interactive setup, this tool creates a robust foundation with modern best practices, allowing you to focus on building unique features.

✨ Key Features

• 🤖 AI-Powered Planning: Describe your project in plain English and let the AI Planner select the best blueprint, frameworks, and structure for you.
• 💬 Interactive & Chat-Based Creation: Choose between a guided interactive CLI that walks you through every option or an AI-driven chat for a "just tell me what you want" experience.
• 🏗️ Multiple Project Blueprints:
  • RAG System: End-to-end Retrieval-Augmented Generation.
  • Multi-Agent System: Sophisticated agentic workflows with LangGraph.
  • Multimodal Chatbot: Chat with text, images, and audio.
  • Core LangChain Application: A modular starting point for any LangChain project.
• 🔌 Flexible Integrations:
  • LLM Providers: OpenAI, Anthropic, Ollama, Mistral.
  • Frameworks: LangChain, LlamaIndex, LangGraph.
  • Vector Stores: Chroma, FAISS.
  • UI Frameworks: Streamlit, Chainlit, FastAPI, Flask, CLI.
• 📦 Modern Python Stack:
  • Type-safe configuration with Pydantic.
  • Easy environment management with .env files.
  • Includes scaffolding for pytest and Jupyter notebooks.
  • Automatic requirements.txt and README.md generation for each project.
• 🔄 Project Updating: Keep your projects current with the ai-bootstrap update command, which seamlessly pulls in the latest template improvements.

---

📋 Prerequisites

Before you begin, ensure you have the following:

• Python 3.9+
• An API key for your chosen LLM provider: OpenAI, Anthropic, or Mistral.
• (Optional) Ollama for running local LLMs.
• (Optional) Docker & VS Code Dev Containers for a sandboxed development environment.

---

🛠️ Installation

Install the tool directly from PyPI:

pip install ai-bootstrap

---

🎉 Getting Started

Creating your first production-ready AI application is just a few commands away.

1. Set Up Your API Keys

Create a .env file in your working directory to store your API keys. The tool will automatically load them.

# .env file
OPENAI_API_KEY="sk-..."
# ANTHROPIC_API_KEY="sk-..."
# MISTRAL_API_KEY="..."
# TAVILY_API_KEY="..." # Optional, for web search tools

2. Create Your First Project (AI-Powered)

This is the recommended and fastest way to get started. Use the --chat flag to let the AI Planner configure your project.

ai-bootstrap create --chat

The tool will prompt you for a short description. For example:

"I want to build a chatbot that can answer questions about multiple PDF documents using a web interface."

The AI Planner will analyze this, generate a project plan, ask for your confirmation, and then scaffold the entire project.

3. Create a Project (Interactive Mode)

If you prefer to make every decision yourself, run the command without any flags.

ai-bootstrap create

This will launch a guided interactive prompt that walks you through selecting a blueprint, UI framework, LLM provider, and other options.

---

📚 Core Commands

Here is a summary of the available commands. For a detailed view, run ai-bootstrap --help.

Command	Description	Example
create	Create a new AI project using either interactive or AI-driven modes.	ai-bootstrap create --chat
update	Update an existing project with the latest template changes.	ai-bootstrap update (run inside project dir)
list-blueprints	Display a table of all available project blueprints and their features.	ai-bootstrap list-blueprints
test-ai-planner	Test the AI Planner with a project description without creating files.	ai-bootstrap test-ai-planner --description "A multi-agent researcher"
help-table	Show a detailed table of all CLI commands and their arguments.	ai-bootstrap help-table

---

🗂️ Project Blueprints

AI Bootstrap provides several battle-tested blueprints for common AI application patterns.

1. RAG System (rag)

   • Description: A complete Retrieval-Augmented Generation system for question-answering over your documents.
   • Tech: LangChain/LlamaIndex, Chroma/FAISS, Streamlit/Chainlit/FastAPI/CLI.

2. Multi-Agent System (multi-agent)

   • Description: An advanced system with multiple AI agents collaborating to solve complex tasks, orchestrated by LangGraph.
   • Tech: LangGraph, Supervisor Pattern, State Management, In-memory/Redis/Postgres memory.

3. Multimodal Chatbot (multimodal-chatbot)

   • Description: A chatbot capable of processing and responding with text, images, and audio.
   • Tech: Chainlit/Streamlit/Flask UI, OpenAI/Anthropic multimodal models.

4. Core LangChain Application (core-langchain)

   • Description: A modular and scalable foundation for building custom LangChain applications.
   • Tech: Custom Chains, Prompt Management, Tool Integration, Streamlit/FastAPI/CLI.

---

📝 Roadmap & Future Plans

This project is under active development. Here are some of the features planned for upcoming releases:

Template Enhancements

• More advanced agent workflows and memory options in multi-agent systems.
• Additional vector store integrations (e.g., Pinecone, Weaviate).
• Enhanced Docker and deployment templates for all blueprints.
• More robust test and notebook scaffolding.

CLI & Application Features

• Project validation and linting before generation.
• Support for custom, user-defined blueprints.
• Richer AI Planner explanations and reasoning in the CLI.
• Automatic virtual environment creation and .env population.

---

🤝 Contributing

Contributions are welcome! If you have ideas for new features, blueprints, or improvements, please open an issue or submit a pull request on our GitHub repository.

📜 License

This project is licensed under the MIT License. See the LICENSE file for details.