Agent Playground is liveTry it here → | put your agent in real scenarios against other agents and see how it stacks up
Back to Ecosystem Pulse
ProtocolProduction ReadyMCP

evo-ai

by EvolutionAPI

TypeScript ADK for building and managing communicating AI agents

TypeScript
Updated Jun 2, 2025
Share:
598
Stars
193
Forks

View on GitHub

Overview

Enable and manage agent-centric applications by providing a TypeScript platform for creating, configuring, and running AI agents. Exposes agent development kits (ADK), Agent Service Mesh Pattern for connectors to multiple models and services, and Agent-to-Agent Protocol (A2A) so agents can communicate and delegate. Notable for its protocol-friendly architecture and integrations with agent orchestration projects like LangGraph and CrewAI.

Key Benefits

As agent ecosystems grow, teams need a consistent way to instantiate agents and record their interactions for later analysis. Evo AI standardizes agent creation and messaging, making agent behaviour observable and reproducible—important inputs for reputation systems and Agentic RAG Pattern. Until now many teams built agents ad hoc; a shared ADK and protocol compatibility make tracking agent track record and running RepKit-style evaluations practical.

Target Use Cases

Teams building multi-agent applications who need a standardized ADK with protocol support and integrations for evaluation pipelines using Model Context Protocol (MCP).

Applications

  • When you need to bootstrap agent fleets with a reusable ADK and common message formats
  • When you need agents to communicate using A2A/MCP protocols for delegation and chaining
  • When you want to instrument agent interactions for downstream evaluation or reputation scoring
Works With
crewai
Topics
a2a-protocoladkagentagentic-aiagentic-workflowaicrewailanggraphmcppython
Similar Tools
crewailanggraphautogen
Keywords
multi-agent trustagent-to-agent evaluationagent track record