Tooning: AI Codebase Chat

Tooning is an industry-grade codebase analyzer and semantic context engine. It bridges the gap between your local source code and Large Language Models (LLMs) by converting complex project hierarchies into high-density TOON payloads.

🔥 Key Features

• 🚀 Concurrent Industrial Scanner: High-performance multi-threaded scanning using fast-glob and p-limit for instant indexing of massive monorepos.
• 🧠 Universal Context Engine: Environment-agnostic core logic that runs identically in VS Code or any standalone terminal.
• 💬 Interactive CLI REPL: Start a context-aware conversation with your codebase directly from your shell with toon chat.
• 📄 Multimodal Native: Full support for PDFs and Excel sheets; documentation is indexed and prioritized alongside source code.
• 🛡️ Security Gates: Automatic blocking of sensitive files (.env, .pem, .key) via industry-standard patterns.

🛠️ Usage

Using the CLI (toon)

Tooning is a standalone powerhouse for terminal enthusiasts:

# Initialize your configuration (Mistral, OpenAI, Anthropic, Ollama)
toon config set provider mistral
toon config set apiKey YOUR_KEY

# Start an interactive, context-aware chat
toon chat

# Generate a one-off TOON payload for external tools (Cursor/Claude)
toon encode . -q "How is concurrency handled?"

Within VS Code

1. Install from the Marketplace.
2. Use the Tooning panel in the Activity Bar for a rich, integrated chat experience.

🏛️ Industry-Grade Foundation

Built for production reliability:

• IConfiguration Abstraction: Shared settings lifecycle across extension and CLI.
• High-Fidelity Symbols: Fallback structural extraction for 10+ professional languages (Go, Rust, Python, C++, etc.).
• Token Efficiency: Two-pass priority encoding to maximize semantic value within context windows.

⚖️ License

Distributed under the MIT License. See LICENSE for details.

Built with ❤️ for the AI-first engineering community.