MarceloClaro/AgentTools ? reverse-engineered prompt
Reverse engineered prompt
Build me a Wolfram Language toolkit that lets AI coding assistants connect to Wolfram through MCP.
I want it to include ready to use server setups for general computation, Wolfram Alpha queries, Wolfram Language work, and paclet development, plus simple install and uninstall helpers for common clients like Claude Desktop, Cursor, VS Code, Claude Code, and the other supported ones. It should let the agent evaluate Wolfram code, search Wolfram documentation and Wolfram Alpha results, read and write notebooks, inspect symbols, run tests, and support project level roots so paths resolve correctly inside a client project.
Please also make it possible for third party paclets to add their own MCP tools and prompts, and include the built in agent skills for Wolfram Language, Wolfram Alpha, notebooks, paclets, and debugging. If needed, look up the current MCP and Wolfram docs online so the client setup and defaults are accurate. Include Docker support too so people can run the server without a local Wolfram Engine.
Are you gonna build this?
make sure you review the code using coderabbit