EREBOTS 4.0 is an open-source hybrid multi-agent framework for conversational assistants. Language-model agents plan and route work, deterministic MCP tools compute exact results, and specialist agents communicate through A2A. The framework adds agents by URL configuration and manages identity, consented user context, task states, and traces. Its A2A server passed 97 official compatibility-kit requirements.

Provides a concrete reference architecture for teams combining language-model routing, deterministic tools, and interoperable specialist agents while preserving task state and traces.

EMMA represents a multi-agent configuration—routing memory, specialist prompts, and procedural skills—as structured episodic memory. A reflector assigns diagnosed failures to update owners, while backbone-independent guards constrain accepted edits. Across three benchmarks, five backbones, and four baselines, the paper reports top overall results and fewer training executions.

The update-ownership and admissibility rules address a recurring problem in agent evolution: changes that drift toward backbone preferences rather than observed failures.