Skill Detail

Compose router, evaluator, and handoff workflows over MCP servers with mcp-agent

Connect LLMs to MCP servers through composable patterns like router, evaluator-optimizer, and orchestrator flows without hand-managing server lifecycles.

Templates & WorkflowsMCP
Templates & Workflows MCP Security Reviewed
⭐ 8.3k GitHub stars ⬇ 29.2k/wk npm
INSTALL WITH ANY AGENT
npx skills add agentskillexchange/skills --skill compose-router-evaluator-and-handoff-workflows-over-mcp-servers-with-mcp-agent Copy
Works best when you want a reusable capability, not another fragile one-off prompt.
At a glance
Tools required
Python runtime, MCP servers, LLM API key, optional Temporal for durable workflows
Install & setup
Install mcp-agent from the upstream package, configure your MCP servers and model credentials, then compose the desired workflow pattern from the quickstart examples.
Author
LastMile AI
Publisher
Open Source Project
Last updated
Apr 19, 2026
Quick brief

Use mcp-agent when the job is to wire an LLM to one or more MCP servers through explicit workflow patterns such as router, map-reduce, evaluator-optimizer, or orchestrator handoffs. The upstream project is centered on MCP-native composition and on managing MCP server connections so the operator can focus on the workflow itself.

How it works

What this skill actually does

Invoke this instead of using raw MCP clients or hand-rolled glue code when you need reusable MCP workflow patterns with clearer boundaries. The scope boundary is specific: this entry is about composing agent workflows over MCP servers. That keeps it skill-shaped and prevents it from collapsing into a generic framework listing.