Capability-based discovery
Tools advertise what they do; agents resolve to whichever endpoint serves that capability at runtime. Add, scale, or replace without rewiring callers.
Tools advertise what they do; agents resolve to whichever endpoint serves that capability at runtime. Add, scale, or replace without rewiring callers.
No central conductor. Each agent reads the shared registry, resolves capabilities, and calls peers directly over HTTP/REST.
Switch between OpenAI, Anthropic, Gemini, Groq, or self-hosted models (Ollama, vLLM, llama.cpp) without changing agent code.
Ordinary Deployments and Services. No proprietary CRDs, no required sidecars, no control plane. ~15–44MB Go containers; OpenTelemetry, circuit breakers, semantic retry built in.