SuperAgent
Orchestrates multiple AI coding agents (Codex, Gemini, Continue) in parallel across different project directories with 16 specialized programming personas like backend architects and security engineers.
Orchestrates Codex and Gemini CLI agents with parallel execution across multiple working directories, enabling delegation of complex tasks to 16 specialized programming personas including backend architects, security engineers, and Python experts for simultaneous multi-project development workflows.
What it does
- Run multiple AI coding agents concurrently
- Execute tasks across different project directories
- Delegate to specialized programming personas
- List available specialized agents
- Manage parallel task execution with timeouts
Best for
About SuperAgent
SuperAgent is a community-built MCP server published by superclaude-org that provides AI assistants with tools and capabilities via the Model Context Protocol. SuperAgent is artificial intelligence development software that orchestrates AI agents for efficient, parallel software It is categorized under ai ml, developer tools. This server exposes 4 tools that AI clients can invoke during conversations and coding sessions.
How to install
You can install SuperAgent in your AI client of choice. Use the install panel on this page to get one-click setup for Cursor, Claude Desktop, VS Code, and other MCP-compatible clients. This server runs locally on your machine via the stdio transport.
License
SuperAgent is released under the MIT license. This is a permissive open-source license, meaning you can freely use, modify, and distribute the software.
Tools (4)
Run Codex CLI agent with parallel execution. Supports multiple tasks concurrently. Use 'workingDirectory' to access different project folders. Codex has full system access. Use 'agent' parameter to invoke a specific agent (run 'list-agents' to see available agents).
Run Gemini CLI agent with parallel execution. Supports multiple tasks concurrently. Use 'workingDirectory' to access different project folders. Auto-approves all actions (YOLO mode). Use 'agent' parameter to invoke a specific agent (run 'list-agents' to see available agents).
Run Continue CLI agent with parallel execution. Supports multiple tasks concurrently. Use 'workingDirectory' to access different project folders. Requires CONTINUE_CONFIG_PATH environment variable to be set. Use 'agent' parameter to invoke a specific agent (run 'list-agents' to see available agents).
List all available specialized agents for use with codex, gemini, and continue tools Available specialized agents: - backend-architect: Design reliable backend systems with focus on data integrity, security, and fault tolerance - business-panel-experts: Multi-expert business strategy panel synthesizing Christensen, Porter, Drucker, Godin, Kim & Mauborgne, Collins, Taleb, Meadows, and Doumont; supports sequential, debate, and Socratic modes. - deep-research-agent: Specialist for comprehensive research with adaptive strategies and intelligent exploration - devops-architect: Automate infrastructure and deployment processes with focus on reliability and observability - frontend-architect: Create accessible, performant user interfaces with focus on user experience and modern frameworks - general-purpose: General-purpose agent for researching complex questions, searching for code, and executing multi-step tasks with thorough analysis - learning-guide: Teach programming concepts and explain code with focus on understanding through progressive learning and practical examples - performance-engineer: Optimize system performance through measurement-driven analysis and bottleneck elimination - python-expert: Deliver production-ready, secure, high-performance Python code following SOLID principles and modern best practices - quality-engineer: Ensure software quality through comprehensive testing strategies and systematic edge case detection - refactoring-expert: Improve code quality and reduce technical debt through systematic refactoring and clean code principles - requirements-analyst: Transform ambiguous project ideas into concrete specifications through systematic requirements discovery and structured analysis - root-cause-analyst: Systematically investigate complex problems to identify underlying causes through evidence-based analysis and hypothesis testing - security-engineer: Identify security vulnerabilities and ensure compliance with security standards and best practices - socratic-mentor: Educational guide specializing in Socratic method for programming knowledge with focus on discovery learning through strategic questioning - system-architect: Design scalable system architecture with focus on maintainability and long-term technical decisions - technical-writer: Create clear, comprehensive technical documentation tailored to specific audiences with focus on usability and accessibility Use the 'agent' parameter to invoke a specific agent.
SuperAgent MCP
SuperAgent is a Model Context Protocol (MCP) server that bridges MCP-compatible clients with the Codex, Gemini, and Continue CLI agents. It lets you fan out multiple CLI tasks in parallel, reuse curated system prompts, and surface results back to your client in a single, structured response.
Key Features
- Unified agent runner: expose the Codex, Gemini, and Continue CLIs as MCP tools without additional wrappers.
- Parallel execution: run multiple prompts concurrently with per-task timeouts and basic result formatting.
- Agent library sync: ship ready-made system prompts that install into
~/.superagent/agentsand can be extended locally. - Tool discovery: query the bundled
list-agentstool to see which specialized agents are available at runtime.
Installation
npm install @superclaude-org/superagent
The postinstall script seeds any missing agent definition files under ~/.superagent/agents so they are immediately
Once connected, the client will discover these tools:
| Tool | Purpose | Notable arguments |
|---|---|---|
codex | Run one or many Codex CLI tasks in parallel. | inputs[] (prompt list), concurrency, workingDirectory, agent, extraArgs, timeoutMs |
gemini | Run Gemini CLI tasks with auto-approval enabled. | inputs[], concurrency, workingDirectory, agent, timeoutMs |
continue | Run Continue CLI tasks with headless mode. Requires CONTINUE_CONFIG_PATH environment variable. | inputs[], concurrency, workingDirectory, agent, timeoutMs |
list-agents | List the specialized agents available to all tools. | (none) |
Invoking a tool
{
"tool": "codex",
"arguments": {
"concurrency": 2,
"inputs": [
{ "prompt": "Run unit tests", "workingDirectory": "/path/to/app" },
{ "prompt": "Summarize latest git changes", "agent": "technical-writer" }
]
}
}
Agent Management
- Agent definitions are Markdown files with frontmatter. You can edit or add new files in
~/.superagent/agents. - The
list-agentstool shows each agent's name and description so you can supply theagentfield when invokingcodex,gemini, orcontinue. - Files shipped with the package are copied only if they do not already exist, preserving local customizations.
Continue Configuration
To use the continue tool, set the CONTINUE_CONFIG_PATH environment variable to point to your Continue config file:
export CONTINUE_CONFIG_PATH="/path/to/.continue/config.yaml"
This can be set in your MCP client configuration's env section.
Development
npm install
npm run build
npm start # runs the compiled server
npm run dev # runs the TypeScript entrypoint with ts-node
Requires Node.js 18 or newer.
Alternatives
Related Skills
Browse all skillsUI design system toolkit for Senior UI Designer including design token generation, component documentation, responsive design calculations, and developer handoff tools. Use for creating design systems, maintaining visual consistency, and facilitating design-dev collaboration.
Answer questions about the AI SDK and help build AI-powered features. Use when developers: (1) Ask about AI SDK functions like generateText, streamText, ToolLoopAgent, embed, or tools, (2) Want to build AI agents, chatbots, RAG systems, or text generation features, (3) Have questions about AI providers (OpenAI, Anthropic, Google, etc.), streaming, tool calling, structured output, or embeddings, (4) Use React hooks like useChat or useCompletion. Triggers on: "AI SDK", "Vercel AI SDK", "generateText", "streamText", "add AI to my app", "build an agent", "tool calling", "structured output", "useChat".
Master API documentation with OpenAPI 3.1, AI-powered tools, and modern developer experience practices. Create interactive docs, generate SDKs, and build comprehensive developer portals. Use PROACTIVELY for API documentation or developer portal creation.
Use when working with the OpenAI API (Responses API) or OpenAI platform features (tools, streaming, Realtime API, auth, models, rate limits, MCP) and you need authoritative, up-to-date documentation (schemas, examples, limits, edge cases). Prefer the OpenAI Developer Documentation MCP server tools when available; otherwise guide the user to enable `openaiDeveloperDocs`.
Guide for building TypeScript CLIs with Bun. Use when creating command-line tools, adding subcommands to existing CLIs, or building developer tooling. Covers argument parsing, subcommand patterns, output formatting, and distribution.
Integrate Vercel AI SDK applications with You.com tools (web search, AI agent, content extraction). Use when developer mentions AI SDK, Vercel AI SDK, generateText, streamText, or You.com integration with AI SDK.