$ loading_
Delegate tasks from an orchestrator agent to multiple CLI-based sub-agents.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "subway-mcp-server" yet — see the docs or source repo.
Use subway-mcp-server to delegate the task "add a logging middleware to a Node.js project" to the Codex sub-agent and return a summary of the result.
A result summary after handing the task to the chosen sub-agent, showing the task was delegated and its output returned.
With subway-mcp-server, send "explain the cause of an error" to Claude CLI and "generate a suggested fix patch" to Codex, then organize both outputs.
Separate outputs from multiple sub-agents plus a combined summary for follow-up work.
Use subway-mcp-server to batch-delegate a set of tasks through non-interactive CLI commands to available sub-agents and return each task status.
Delegation status for each task and returned results from sub-agents, suitable for automation workflows.
When a developer uses one orchestrator agent to manage work, this MCP tool can send coding or analysis tasks to sub-agents such as Codex, Claude, or Antigravity CLI. It helps break complex workflows into delegated steps.
In automation flows that rely on non-interactive commands, teams can use this local MCP server to forward tasks to sub-agents. It fits scripted or pipeline-style agent orchestration.
It is a local MCP server that exposes delegation tools to an orchestrator agent. The orchestrator can assign tasks to sub-agents such as Codex, Claude, and Antigravity CLI through non-interactive CLI commands.
Based on the description, you need an MCP-capable orchestrator environment and the relevant sub-agent CLIs. For exact runtime, installation, or configuration requirements, see the source repository.
Its focus is not completing tasks alone, but letting one orchestrator delegate work across multiple sub-agents. That makes it better suited for multi-agent collaboration and automated distribution.
Delegate Claude Code subagents to local or external model backends seamlessly.
Orchestrate multiple AI CLIs with retries, failover, and resilient task routing.
Delegate coding tasks to Codex via MCP with security and result checks.
Route local coding agent CLIs with validation, retries, and capability-based dispatch.
Connect coding agents to multiple models for in-chat consultation and code review.
Turn your AI client into a coding hub with execution, memory, and sub-agents.