Show HN: Bot-inator – Mac agent that writes its own tools, runs them sandboxed A developer released Bot-inator, an MIT-licensed, local-first macOS agent that watches system activity and offers to write its own tools, which run only after user approval inside a short-lived child process under macOS sandbox-exec. The app ships as a signed and notarized Apple Silicon download (macOS 14 or later, about 540 MB) with a Python FastAPI daemon on 127.0.0.1:8000 and a SwiftUI menu bar UI talking over a WebSocket, and runs models locally via MLX, with an optional Groq API key for cloud inference. Generated tools pass static checks, a second model pass comparing code to intent, and a dry run in a throwaway sandbox before the user reads the full source, and each installed tool is versioned in its own git repo with repair offers and diffs for failing tools. A local-first macOS agent that watches what you do and offers to help. Nothing runs without your approval. Download for macOS https://github.com/Bforsyth1234/bot-inator/releases/latest Apple Silicon, macOS 14 or later, signed and notarized or build from source build-from-source . It notices activity on your Mac new files in Downloads, incoming mail and iMessages, upcoming meetings, app switches , asks an LLM "how could I help here?" , and offers an action, or a small tool it writes itself, as a card in the menu bar. bot-inator is not limited to the tools it ships with. When it spots something you keep doing by hand, or something you asked for in chat that it couldn't do, it offers to build a tool for it: 1. It offers the idea first. A card says what the tool would do; nothing is written until you approve. 2. It drafts and checks the code. Static checks, a second model pass that compares the code with the idea, and a dry run in a throwaway sandbox with a fake home folder that closes the same folders as the real sandbox. Problems go back to the model to fix before you see anything. 3. You read the full source and approve it, edit it or throw it away. 4. It runs sandboxed. Each call starts a short-lived child under macOS sandbox-exec , and still asks before it runs. 5. It stays fixable. Every installed tool is versioned in its own git repo, and a tool that keeps failing gets a repair offer with a diff. Ways in besides the automatic offers: ask in Chat, or use Watch me to do a task once while it records the steps. What you can check for yourself, in this repo: - Every action is a card. Tools that change something move a file, create a reminder or event, draft an email ask first; only read-only tools skip the card. See READ ONLY TOOL NAMES in agent-daemon/tools/ init .py . - Generated tools are reviewed, then sandboxed. You read the full source before a tool is installed, and each call runs in a short-lived child under macOS sandbox-exec How generated tools run how-generated-tools-run . - Local by default. Models run on your Mac with MLX. A cloud provider is opt-in, and email and message text stay on the Mac unless you also allow them to leave. Settings → Privacy ledger lists what left today. MIT licensed LICENSE https://github.com/Bforsyth1234/bot-inator/blob/main/LICENSE . | Part | What it is | |---|---| | agent-daemon/ | Python service FastAPI on 127.0.0.1:8000 . Listens to system events, runs the LLM, holds the tools. | | MenuBarUI/ | SwiftUI menu bar app. Shows the agent's thoughts, asks for approvals, chat, tool manager, settings. | The two talk over a WebSocket /ws/stream plus a small HTTP API. See ARCHITECTURE.md https://github.com/Bforsyth1234/bot-inator/blob/main/ARCHITECTURE.md for diagrams and LIVING SPEC.md https://github.com/Bforsyth1234/bot-inator/blob/main/LIVING SPEC.md for the full protocol and design history. - macOS 14+ on Apple Silicon local models use MLX - To build from source only: Python 3.11+ developed and tested on 3.14 and the Xcode command line tools xcode-select --install . The download needs neither. - Optional: a Groq API key https://console.groq.com/keys to run models in the cloud instead of on your Mac 1. Download bot-inator-