This article was published on the Ippon Technologies blog (in French).
The Model Context Protocol (MCP) allows you to expose internal tools to an AI agent directly in your IDE — without IDE-specific plugins, without ad hoc integrations. Instead of giving suggestions, the agent can execute real actions on your codebase.
In this article, I walk through our implementation: a Rust CLI exposed via a TypeScript MCP server, the design principles that make tools genuinely useful for an agent, and the pitfalls to avoid.
- Why MCP instead of a classic VSCode plugin
- How to structure outputs so an agent can make decisions
- Concrete implementation: Rust CLI + TypeScript façade
Read the full article on the Ippon blog: Model Context Protocol (MCP): Exposing a Rust CLI to Your IDE