AgentStack
AgentStack is a production-grade multi-agent framework built on Mastra, delivering 50+ enterprise tools, 25+ specialized agents, and A2A/MCP orchestration for scalable AI systems. Focuses on financial intelligence, RAG pipelines, observability, and s
What is AgentStack?
AgentStack is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to agentstack is a production-grade multi-agent framework built on mastra, delivering 50+ enterprise tools, 25+ specialized agents, and a2a/mcp orchestration for scalable ai systems. focuses on financial...
AgentStack is a production-grade multi-agent framework built on Mastra, delivering 50+ enterprise tools, 25+ specialized agents, and A2A/MCP orchestration for scalable AI systems. Focuses on financial intelligence, RAG pipelines, observability, and s
This server falls under the Coding Agents category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- AgentStack is a production-grade multi-agent framework built
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx agentstackConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use AgentStack
AgentStack is a production-grade multi-agent framework built on Mastra that ships 48+ specialized agents, 94+ enterprise tools, and 21 multi-step workflows out of the box. It integrates financial data APIs (Polygon, Finnhub, AlphaVantage), search tools (SerpAPI), a zero-config RAG pipeline with libSQL vector storage, browser automation via Chrome/CDP, and full observability through Langfuse. Teams building complex AI systems can use AgentStack as a foundation for financial intelligence, content creation pipelines, document processing, and scalable A2A/MCP orchestration.
Prerequisites
- Node.js 20.9.0 or later
- A Google Generative AI API key (GOOGLE_GENERATIVE_AI_API_KEY) for Gemini LLM and embeddings
- A Turso database URL or local libSQL file path for persistence
- A SerpAPI key (SERPAPI_API_KEY) for search tools
- A Polygon API key (POLYGON_API_KEY) for financial data tools
Clone the repository
Clone the AgentStack repository and navigate into the project directory.
git clone https://github.com/ssdeanx/AgentStack.git
cd AgentStackInstall dependencies
Install all Node.js dependencies. Node.js 20.9.0 or later is required.
npm installConfigure environment variables
Create a .env file in the project root with your API keys and database configuration. At minimum, set the Gemini key and a database URL.
GOOGLE_GENERATIVE_AI_API_KEY=your-gemini-api-key
TURSO_DATABASE_URL=file:./data/mastra.db
SERPAPI_API_KEY=your-serpapi-key
POLYGON_API_KEY=your-polygon-key
# Optional
OPENAI_API_KEY=your-openai-key
LANGFUSE_BASE_URL=https://cloud.langfuse.comStart the development server
Launch the AgentStack development server. The web interface will be available at http://localhost:3000.
npm run devConfigure Claude Desktop to use AgentStack via MCP
Add AgentStack as an MCP server in your Claude Desktop configuration to expose its agents and tools to Claude.
{
"mcpServers": {
"agentstack": {
"command": "npx",
"args": ["agentstack"],
"env": {
"GOOGLE_GENERATIVE_AI_API_KEY": "your-gemini-api-key",
"TURSO_DATABASE_URL": "file:./data/mastra.db",
"SERPAPI_API_KEY": "your-serpapi-key",
"POLYGON_API_KEY": "your-polygon-key"
}
}
}
}Run tests to verify your setup
Run the test suite to confirm all components are correctly configured. AgentStack ships with 97% test coverage.
npm test
npm run typecheckAgentStack Examples
Client configuration
Claude Desktop configuration for AgentStack with the required API keys.
{
"mcpServers": {
"agentstack": {
"command": "npx",
"args": ["agentstack"],
"env": {
"GOOGLE_GENERATIVE_AI_API_KEY": "your-gemini-api-key",
"TURSO_DATABASE_URL": "file:./data/mastra.db",
"SERPAPI_API_KEY": "your-serpapi-key",
"POLYGON_API_KEY": "your-polygon-key"
}
}
}
}Prompts to try
Example prompts that exercise AgentStack's financial intelligence, RAG, and multi-agent capabilities.
- "Generate a financial intelligence report for AAPL including recent price aggregates, fundamentals, and news sentiment."
- "Search the web for the latest trends in AI agent frameworks and summarize the top 5 findings."
- "Ingest this PDF document into the RAG pipeline and answer questions about its contents."
- "Delegate a content creation workflow: research the topic, draft an article, and return the final output."
- "What specialized agents are available and what can each one do?"Troubleshooting AgentStack
Server fails to start with 'GOOGLE_GENERATIVE_AI_API_KEY is not set'
Ensure your .env file is in the project root and contains GOOGLE_GENERATIVE_AI_API_KEY. The server requires this key at startup. Verify with 'node -e "require(\"dotenv\").config(); console.log(process.env.GOOGLE_GENERATIVE_AI_API_KEY)"'.
libSQL/Turso connection fails on startup
For local development, use TURSO_DATABASE_URL=file:./data/mastra.db and ensure the data/ directory exists (mkdir -p data). For Turso cloud, verify the URL and auth token are correct in your .env file.
Financial data tools return empty results
Confirm POLYGON_API_KEY is set and valid. Polygon's free tier has rate limits and data restrictions. Check your Polygon dashboard for API usage and upgrade to a paid plan if you need real-time data or higher request limits.
Frequently Asked Questions about AgentStack
What is AgentStack?
AgentStack is a Model Context Protocol (MCP) server that agentstack is a production-grade multi-agent framework built on mastra, delivering 50+ enterprise tools, 25+ specialized agents, and a2a/mcp orchestration for scalable ai systems. focuses on financial intelligence, rag pipelines, observability, and s It connects AI assistants to external tools and data sources through a standardized interface.
How do I install AgentStack?
Follow the installation instructions on the AgentStack GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with AgentStack?
AgentStack works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is AgentStack free to use?
Yes, AgentStack is open source and available under the MIT license. You can use it freely in both personal and commercial projects.
AgentStack Alternatives — Similar Coding Agents Servers
Looking for alternatives to AgentStack? Here are other popular coding agents servers you can use with Claude, Cursor, and VS Code.
Dify
★ 142.2kProduction-ready platform for agentic workflow development.
Ruflo
★ 54.0k🌊 The leading agent orchestration platform for Claude. Deploy intelligent multi-agent swarms, coordinate autonomous workflows, and build conversational AI systems. Features enterprise-grade architecture, self-learning swarm intelligence, RAG integrat
Goose
★ 45.7kan open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
Antigravity Awesome Skills
★ 38.3kInstallable GitHub library of 1,400+ agentic skills for Claude Code, Cursor, Codex CLI, Gemini CLI, Antigravity, and more. Includes installer CLI, bundles, workflows, and official/community skill collections.
AgentScope
★ 25.5kBuild and run agents you can see, understand and trust.
Serena
★ 24.5kA coding agent toolkit that provides IDE-like semantic code retrieval and editing tools, enabling LLMs to efficiently navigate and modify codebases using symbol-level operations instead of basic file reading and string replacements.
Browse More Coding Agents MCP Servers
Explore all coding agents servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up AgentStack in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use AgentStack?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.