What it does
Agent Orchestrator MCP lets a user create agents, delegate tasks using trust-based routing, coordinate file access, run sprints and monitor performance. Its listed features include dependency resolution, parallel execution, retry with backoff and progress tracking.
It suits workflows where several agents share one body of work, such as tasks that depend on each other, tasks that can run in parallel, or tasks that need retrying after failure. The publisher states that the server was built with provisions relating to EU AI Act Articles 9, 13, 15, 26 and 50, covering risk management, transparency, bias testing, FRIA support and AI content watermarking where applicable.
To run it, add the server to an MCP client configuration, such as the claude_desktop_config.json file used by Claude Desktop, or install the Python package with pip and start the agent-orchestrator-mcp command, which uses stdio transport. Once configured, an assistant can call tools including create_agent, list_agents and delegate_task.
Limitations and requirements
Requires an MCP client, such as Claude Desktop, configured to launch the server, or a Python environment for installation with pip. The server runs over stdio transport.
Safety
Our team has not reviewed this resource. Only automated rules checked it; resources with high-risk signals are not published. These are signals, not a guarantee that installing it is safe. Try it in a sandbox project first.
- Runs code or changes an assistant's behaviour
Sources
- Official MCP Registry first seen 13 Nov 2026 · last seen 13 Nov 2026 · its page on the source
io.github.CSOAI-ORG/agent-orchestrator-mcp
Agent Orchestrator MCP Server by MEOK AI Labs
v1.0.12As shown on registry.modelcontextprotocol.io when we last read it