Show HN: Toli – Let Claude Code on one computer work with Codex on another (github.com)

Toli, a new tool designed to enable collaboration among AI agents across various platforms, has been introduced to the community. This innovative solution allows AI models, such as Claude Code and Codex, to work together seamlessly within local environments or over the internet. By utilizing the A2A (Agent-to-Agent) protocol, agents can delegate tasks, ask questions, and share results, thereby enhancing their overall functionality and productivity. Toli establishes a daemon that allows users to connect multiple AI agents, each with unique identifiers and tokens, facilitating their interaction and task management through a command-line interface or a text-based user interface. The significance of Toli in the AI/ML landscape lies in its potential to unify different AI models and systems, fostering interoperability and collaboration. By enabling agents to communicate regardless of their operational environments, Toli supports a wide array of applications beyond coding, such as assisting personal assistants in providing accurate responses. The tool's adaptability, security features using Tailscale for encryption, and straightforward installation process make it accessible for developers looking to enhance their AI workflows. With an architecture that promotes scalability and an open-source license, Toli opens new avenues for innovation in AI agent collaboration.