opencode-rust: an open-source AI coding agent in Rust with CLI, TUI, and HTTP server

⚡ The gist in 5 seconds - A project called vakandi/opencode-rust has appeared on GitHub: an open-source AI agent for coding, written in Rust. - Interfaces include a CLI, a TUI, and an HTTP server; agents run locally without Docker. - Key limitation of the source: the repository description does not reveal supported models or the configuration scheme. ### 🔍 What was found The vakandi/opencode-rust repository publishes an open-source AI agent for working with code, implemented in Rust. The project is described as an "open source AI coding agent" and provides three ways to interact right away: a command-line interface (CLI), a text-based interface (TUI), and an HTTP server for local AI agents. It specifically emphasizes that Docker is not required to run it. Judging by the description, the architecture is split into several entry points: the CLI and TUI are for interactive work in the terminal, while the HTTP server is for accessing the agent over the network or integrating with external clients. No details about the list of supported models, how to connect them, or the config format are given in the repository title or description — only a claim of local execution without containerization. ### 💡 Why it matters Skipping Docker is a practical argument for those running agents on their own machine: fewer layers, less container overhead, and an easier stack to deploy on a bare system. The "CLI + TUI + HTTP server" format covers three scenarios at once — manual coding in the terminal, an interactive session, and programmatic integration. The source still lacks specifics on features, licensing, and supported models: the project should be judged by the code in the repository, not by the description alone.