Composer MCP Server

Composer MCP Server is a platform that helps users backtest trading ideas and automate investments. It works with MCP-enabled LLMs like Claude, allowing users to create and validate trading strategies seamlessly. The server supports a variety of indicators such as Relative Strength Index (RSI), Moving Average (MA), and Exponential Moving Average (EMA). Users can build their ideal portfolio with a diverse range of equity and crypto offerings. The Backtesting API provides a fast feedback loop for AI to iterate and validate its hypotheses. Users can monitor the performance of individual strategies and overall accounts, view performance statistics, and analyze their best-performing symphonies. The server also allows users to control their investments by asking AI to analyze and adjust their exposure. Composer MCP Server is compatible with various AI clients, including Claude Desktop, Cursor, and OpenAI Agents. To use the server, users need to install the necessary components, connect their LLM, and create or backtest trading strategies using commands. An API key is required for advanced features, which can be obtained from the Composer website. The server offers a 14-day free trial for new users. For the best experience, it is recommended to use Claude Opus 4 instead of Sonnet, as Opus is much better at tool use. Users should also turn on Claude's Research mode if they need the latest financial data and news. Tools that execute trades or affect funds should only be allowed once and not set to "Always Allow". The server provides a range of tools for creating, backtesting, and managing symphonies, as well as monitoring performance and executing trades. For troubleshooting, logs can be found in specific locations depending on the operating system. The ModelScope MCP Server integrates with ModelScope's ecosystem, providing seamless access to AI models, datasets, apps, papers, and generation capabilities through popular MCP clients. It offers features such as user authentication, AI image generation, model search, research paper search, and Gradio API integration. To use the ModelScope MCP Server, users need to obtain an API token from ModelScope and add the configuration to their MCP client's configuration file. The server can be run locally or using a pre-built Docker image. The project welcomes contributions and has guidelines for code quality and testing. Overall, both the Composer MCP Server and the ModelScope MCP Server provide powerful tools for integrating AI capabilities with trading and research applications.