Show HN: Myco – coordinate Claude and DeepSeek and other LLMs in one agent swarm (github.com)

A new coordination protocol and Python daemon called Myco has been introduced to address challenges faced by developers running multiple AI agent sessions in parallel, such as conflicting actions and stale information. Myco allows for simultaneous operation of different AI models like Claude and DeepSeek, providing a shared, filtered view of each agent's actions without a central orchestrator. The protocol, inspired by mycelium's underground networks, enables persistent sessions to communicate effectively while maintaining clean contexts for each agent, improving productivity and minimizing chaos in collaborative environments. The significance of Myco lies in its capability to facilitate human-in-the-loop coordination among various AI models without vendor lock-in, allowing developers to choose the best tools for specific tasks. This approach improves resilience, as agents can continue to function even if one provider experiences downtime. With a straightforward integration process and support for multiple platforms, Myco represents an important step in enhancing the efficiency of AI/ML workflows, enabling a seamless exchange of information between independent agents while reducing duplication of effort and potential errors in complex projects.