awesome-finance
A curated list of MCP servers for AI finance agents
claude mcp add --transport stdio blockrunai-awesome-finance-mcp node server.js \ --env PORT="Port to run the MCP server on" \ --env DB_CONNECTION="Database connection string (if required)"
How to use
Awesome Finance MCP provides a collection of finance-focused AI workflows and data tools that you can query or chain together as part of an autonomous agent. The server exposes multiple workflows for stock and crypto data access, trading execution hooks, and analytical utilities that can be composed into higher-level tasks. Typical usage involves starting the server, then invoking the available MCP endpoints or integrating the server into your agent’s prompt-based toolset to fetch market data, run analyses, or simulate trades. The included capabilities are designed to work with common MCP client tooling, enabling you to plug in real-time price feeds, indicators, backtests, or on-chain analytics as part of your agent’s decision loop.
Once running, you can discover available skills or workflows via the server’s API or documentation accompanying the repository. These tools are intended to be combined in sequences—such as obtaining current data, performing a technical analysis, and then triggering a simulated or actual trade—allowing finance-focused agents to operate with modular, testable components.
How to install
Prerequisites:
- Node.js (LTS) and npm installed on your system
- Git installed
- Optional: a compatible database or environment variables if the MCP server expects persistence
Installation steps:
-
Clone the repository: git clone https://github.com/your-org/awesome-finance.git cd awesome-finance
-
Install dependencies: npm install
-
Configure environment (example): PORT=3000 DB_CONNECTION=postgres://user:pass@localhost:5432/awesomefinancedb
On Windows use set PORT=3000 first, or use a .env file and a library like dotenv
-
Run the server: npm run start
If your package.json uses a different script, use the appropriate command, e.g., node server.js
-
Verify the server is running by visiting http://localhost:3000 (or the configured port) or by calling a health/check endpoint if provided.
Additional notes
Notes and tips:
- If you’re integrating with real trading providers, ensure you have the proper API keys or credentials configured as environment variables as required by the workflows.
- Some workflows may require external data providers or websocket streams; ensure network access and any necessary subscriptions are in place.
- Start with read-only or simulated modes if available to validate the workflows before enabling live trading capabilities.
- Use the server’s health or status endpoint to monitor uptime and performance in production.
- If you encounter port conflicts, reconfigure PORT to an available port and restart.
- Review any rate limits or pricing notes for included data providers to avoid unexpected costs.
- For local development, consider adding a .env file and using a dedicated development database to prevent data leakage between environments.
Related MCP Servers
Awesome
Awesome ModelContextProtocol resources - A curated list of MCP resources
hyperliquid
🚀 MCP Server para Hyperliquid DEX - Trade com Claude usando linguagem natural. Desenvolvido por Caio Vicentino para as comunidades Yield Hacker, Renda Cripto e Cultura Builder
WAIaaS
Wallet-as-a-Service for all AI agents in the world
web3 -hub
The definitive open-source registry of Model Context Protocol (MCP) servers for Web3, blockchain, and decentralized applications. 130+ chains, DeFi, NFTs, analytics, and identity tools
philidor
MCP server for DeFi vault risk analytics. Search 700+ vaults, compare risk scores, analyze protocols. No API key needed.
pumpclaw
Free token launcher for AI agents on Base. Uniswap V4. 80% creator fees. LP locked forever. Deploy via Farcaster (@clawd), CLI, or smart contract.