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
ToolProduction ReadyMCP

yao

by YaoApp

Self-hosted agent workspaces and task boards across devices

Go
Updated Sep 5, 2026
Share:
7.9k
Stars
704
Forks

View on GitHub

Overview

Unifies agent workspaces, tasks, and boards across devices with a self-hosted Go backend and web/mobile clients. Exposes APIs and a board-first UI so teams can create, run, and monitor agents and skills in one place. Notable for local-first, self-hosted operation and integrations that let you attach LLM providers and agent runtimes to persistent workspaces.

Key Benefits

As teams run many specialized agents, having a single place to track their tasks, state, and outputs makes auditing and reputation signals practical. YaoApp turns informal agent interactions into persistent records and workspaces, enabling correlation of behavior over time — a basic building block for agent track records and reproducible A2A evaluation. Until agents log their activities and results, meaningful trust or continuous evaluation is hard to achieve; YaoApp helps make that logging and organization standard practice. A2A evaluation

Target Use Cases

Teams and developers who want a self-hosted workspace to author, run, and track multiple agents and their tasks in production-like environments.

Use Cases

  • Organize and track agent-led tasks across teams with persistent workspaces
  • Attach LLM providers to reproducible agent runs for pre-production testing
  • Log agent interactions and outputs to build an audit trail for reputation and debugging
Works With
openaianthropicclaude
Topics
agent-harnessagent-orchestrationagent-skillsagent-workspacesai-agentsai-nativeclaude-codecoding-agentdeepseekdeepseek-harness+7 more
Similar Tools
autogenlangchain
Keywords
multi-agent orchestrationagent workspacesagent track recordself-hosted