Building a safer desktop AI agent: Rust, Tauri, and explicit tool approvals A developer has released Praxi-Bot, an Apache-2.0 open-source desktop AI agent built on a Rust daemon (Tokio and Axum) with a Tauri and React front end, designed so that LLM tool use is gated by explicit user approvals. The default 'Approve destructive' policy lets read-only actions proceed while file modifications, desktop input, command execution and messaging require approval, with financial actions always gated and critical credential and system paths blocked. Node agents connect outward to the daemon, so controlled machines need no inbound ports, and the project remains alpha software supporting Gemini, Anthropic, OpenAI-compatible endpoints and local Ollama. I'm working on Praxi-Bot https://github.com/emreocell/Praxi-Bot , an Apache-2.0 open-source AI agent built around a Rust daemon and a Tauri desktop app. One architectural choice keeps coming up in this project: how do you let an LLM use real tools without turning every suggestion into an unrestricted system action? Connecting a model to a shell or browser is the easy part. The difficult part is deciding which operations require approval, making those decisions visible, and retaining a useful user experience. Praxi-Bot currently separates the desktop application from its Rust-based background daemon. The daemon uses Tokio and Axum, while the desktop app uses Tauri and React. Node agents connect outward to the daemon, avoiding a requirement to open inbound ports on controlled machines. The application remains alpha software : APIs and internals may change, and the latest prerelease is the supported version. The Windows first-run wizard includes a safety configuration. The default is Approve destructive : read-only interactions can proceed, while operations that modify files, type or click on the desktop, run commands, or send messages ask for approval. Financial actions always require approval, and critical credential/system paths are blocked. That is a practical design constraint: an agent that can observe a desktop isn't automatically entitled to change it. It also changes the UI requirements. Users need understandable permission prompts that describe the requested action, not a vague "Allow agent?" dialog. Praxi-Bot supports multiple model options, including Gemini, Anthropic, OpenAI-compatible endpoints and local Ollama. A user can choose their own provider, while the app can focus on the permission model and tools instead of hard-wiring itself to one API. The repository includes documentation for onboarding, workflows, skills, Telegram pairing, and the CLI. Planned features are marked as such in the guide rather than presented as shipped. The hard questions are still open for discussion: If you work with Rust, React/TypeScript, accessibility, testing, or agent security, I'd welcome both critical feedback and code contributions. There are contribution guidelines https://github.com/emreocell/Praxi-Bot/blob/main/CONTRIBUTING.md , starter tasks https://github.com/emreocell/Praxi-Bot/blob/main/docs/contributing/starter-issues.md , and a good first issue label https://github.com/emreocell/Praxi-Bot/labels/good%20first%20issue . AI-assisted PRs are welcome when contributors understand and verify their changes. Source: https://github.com/emreocell/Praxi-Bot https://github.com/emreocell/Praxi-Bot What permission model has worked best in a real agent application you've built or used?