Sub-Agents MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| AGENTS_DIR | Yes | Absolute path to your agents folder where .md or .txt agent definitions are stored. | |
| AGENT_TYPE | Yes | The execution engine to use: 'cursor' (cursor-agent), 'claude' (claude CLI), 'gemini' (gemini CLI), or 'codex' (codex CLI). | |
| SESSION_DIR | No | Absolute path to the directory where session history files are stored. | |
| SESSION_ENABLED | No | Set to 'true' to enable session management for context continuity between executions. | false |
| EXECUTION_TIMEOUT_MS | No | How long agents can run before timing out in milliseconds. Default is 5 minutes (300000), max is 10 minutes. | 300000 |
| SESSION_RETENTION_DAYS | No | How long to keep session files based on last modification time in days. | 1 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {} |
| resources | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| run_agentA | Delegate complex, multi-step, or specialized tasks to an autonomous agent for independent execution with dedicated context (e.g., refactoring across multiple files, fixing all test failures, systematic codebase analysis, batch operations). Returns session_id in response metadata - reuse it in subsequent calls to maintain conversation context continuity across multiple agent executions. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Agent List | List of available Claude Code sub-agents |
| Agent: README | Sub-Agents MCP Server |
TDQS
Scored across 1 tool
With only one tool, there is no possibility of confusion or overlap between tools. The single tool 'run_agent' has a clearly defined and distinct purpose of delegating complex tasks to autonomous agents.
The naming pattern cannot be inconsistent with only one tool. The tool name 'run_agent' follows a clear verb_noun pattern, which would be consistent if more tools existed.
A single tool is too few for a server named 'Sub-Agents MCP', which implies functionality for managing or interacting with multiple agents. The scope suggests operations like listing agents, checking status, or controlling execution, but only delegation is provided.
The tool surface is severely incomplete for the implied domain of sub-agent management. While 'run_agent' handles task delegation, there are obvious gaps such as listing available agents, monitoring agent status, terminating agents, or configuring agent parameters, which are essential for a coherent agent management system.