Skip to main content
Glama

MCP-Swarm-Router

An MCP server that lets an LLM host (e.g. Claude Code) plan a large task, then delegate subtasks to local CLI agents based on their specialization — a "swarm router" for tools already installed on your machine.

Roster

id

specialization

CLI assumed

non-interactive invocation

codex

Image generation & heavy tools

OpenAI Codex CLI

codex exec --sandbox workspace-write "<prompt>"

agy

UI/UX design & creative

Google Antigravity CLI (agy, Gemini-based) — unverified, swap via env if wrong

agy -p "<prompt>" --dangerously-skip-permissions

claude-code

Hard coding & logic

Claude Code

claude -p "<prompt>" --output-format text --permission-mode acceptEdits

--sandbox workspace-write / acceptEdits / --dangerously-skip-permissions are there so each CLI can actually write files without blocking on an approval prompt — none of these processes have a TTY, so an unhandled prompt would hang until timeout_ms.

Edit src/registry.ts to add/remove agents or change how each one is invoked.

Related MCP server: delegations-mcp

Tools

  • get_agent_roster — returns the roster above as JSON (id, specialization, description, resolved CLI command). Call this first to decide which agent_name fits a subtask.

  • delegate_task — spawns agent_name's CLI in workspace_path with prompt, waits for it to exit (or hit timeout_ms, default 15 min, max 1 hour), and returns { exitCode, signal, stdout, stderr, timedOut, durationMs }. Output per stream is capped at ~5MB (further output is dropped and truncated is set).

There's also a plan-and-delegate MCP prompt that spells out the intended workflow: call get_agent_roster → break the task into subtasks → confirm the plan for anything large → run delegate_task per subtask → summarize results.

Setup

npm install
npm run build

Each agent's CLI must be installed and reachable on PATH under the name in src/registry.ts (codex, agy, claude). If your local install differs, copy .env.example to .env (or set the vars in your shell / MCP client config) to point at the real binary:

MCP_SWARM_CODEX_CMD=
MCP_SWARM_AGY_CMD=
MCP_SWARM_CLAUDE_CMD=

Spawning goes through cross-spawn rather than a shell, so Windows .cmd/.bat shims (npm global installs) resolve correctly and the prompt text can't be interpreted as shell metacharacters.

Register with an MCP client

{
  "mcpServers": {
    "swarm-router": {
      "command": "node",
      "args": ["C:/KERJAAN/mcp-swarm-router/dist/index.js"]
    }
  }
}

Development

npm run dev    # run src/index.ts directly via tsx
npm run build  # compile to dist/
npm start      # run the compiled server

Available Tools

2 tools
delegate_taskDelegate TaskA

Spawns the named agent's CLI in workspace_path with the given prompt, waits for it to finish, and returns its stdout/stderr/exit code. Use get_agent_roster first to pick the right agent_name.

Inputs: prompt (required), agent_name (required), workspace_path (required), timeout_ms.

get_agent_rosterGet Agent RosterA

Returns the list of local agents available for delegation, each with its id and specialization. Call this before delegate_task to decide which agent_name fits a given subtask.

Inputs: no named parameters.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.3/5.0

    Scored across 2 tools

    Disambiguation5/5

    The two tools have clearly distinct purposes: one provides an inventory of available agents, the other executes a task with a specified agent. There is no overlap or ambiguity between them.

    Naming Consistency5/5

    Both tool names follow a consistent verb_noun pattern (get_agent_roster, delegate_task), using snake_case and clear action verbs. The naming is uniform and predictable.

    Tool Count3/5

    With only 2 tools, the set is on the thin side but still covers the core workflow of a router: discovering agents and delegating to them. It feels slightly sparse for a 'swarm' concept, but each tool is essential.

    Completeness4/5

    The tool pair covers the full delegation lifecycle needed: learn who is available, then execute a task. Missing features like parallel delegation or cancellation are minor gaps for a simple router implementation, and agents can work around them by calling delegate_task repeatedly.

    Maintenance

    ActivityNo data
    ResponsivenessOutdated data

    Related MCP Connectors

    Related MCP Servers

    • F
      license
      Not graded
      quality
      D
      maintenance
      A lightweight MCP server that enhances AI agents with tools for codebase analysis, task delegation to sub-agents, multi-agent coordination through chatrooms, and project todo management.
      -
    • F
      license
      B
      quality
      D
      maintenance
      An MCP server that exposes a library of delegation prompts to orchestrate tasks between a primary LLM and specialized sub-agents. It enables the execution of self-contained, bounded tasks with built-in support for configuration discovery and project-specific delegation libraries.
      3
      -
    • A
      license
      Not graded
      quality
      D
      maintenance
      An MCP server that allows a planning agent to delegate tasks to executor agents (e.g., Claude Code, Aider) with bi-directional communication and real-time log streaming.
      19 npm
      15
      AGPL 3.0
    • A
      license
      A
      quality
      D
      maintenance
      Local MCP server that exposes delegation tools for Codex, Claude, and Antigravity CLI, enabling an orchestrator agent to assign tasks to these sub-agents via non-interactive CLI commands.
      3
      MIT