Show HN: Myco – coordinate Claude and DeepSeek and other LLMs in one agent swarm
Myco is a coordination protocol designed to manage multiple AI agent sessions simultaneously without conflicts. It allows agents like Claude and DeepSeek to share relevant information while maintaining their individual contexts. This tool aims to enhance productivity by reducing chaos in collaborative coding environments.
- ▪Myco enables parallel sessions of AI agents to communicate through a shared, filtered view of their actions.
- ▪The protocol is vendor-agnostic and works with any Anthropic-compatible model.
- ▪Myco addresses the challenges of running multiple agent sessions by preventing conflicts and stale assumptions.
Opening excerpt (first ~120 words) tap to expand
myco 🇧🇷 Leia em português Problem: running multiple Claude Code sessions in parallel produces conflicts, repeated work, and stale assumptions — the agents have no shared awareness. myco is a text-only coordination protocol + tiny Python daemon that gives N parallel agent sessions a shared, filtered view of each other's actions. No central orchestrator. No tool calls. Vendor-agnostic (Claude, DeepSeek, anything Anthropic-compatible). The name comes from the mycelium: an underground fungal network that connects independent trees, silently transporting signals between them. Why If you use Claude Code (or any agent CLI) intensively, you've probably hit the wall: Long sessions degrade attention. Even with a 1M context window, late-conversation precision drops.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.