b-nnett/grok-bot-0.18-reconstructed

This page introduces an external open-source repository. It is not an HDATF product.

What it does

An unofficial reconstruction of the Grok Bot 0.18.0 macOS app in readable TypeScript, with a toolchain to rebuild it. It adds an inference router, MCP tools across providers, local usage tracking and an optional Docker sandbox.

How it helps ATF

Its inference router across Cursor, Claude Code, Codex and OpenRouter could be a loose reference for how Harness hands work to different executors. Because it unofficially rebuilds a shipped app, only its ideas are worth studying.

License

No license No license file found. Without a license, reuse of the code is not permitted by default.

More in this category

• vastsa/PI-Desktop
  PI-Desktop is a local-first desktop app for AI coding agents, built on Electron, a Rust host core and the pi agent harness. Projects, sessions, reviews, previews, model settings, permissions and plugins sit in one workspace, with any configured model.
• xai-org/grok-build
  A terminal-based AI coding agent with a full-screen interface. It reads the codebase, edits files, runs shell commands, searches the web and manages long-running tasks, interactively, headlessly for scripts and CI, or inside editors through the Agent Client Protocol.
• microsoft/tgrep
  tgrep builds a trigram index of a codebase and runs a server that watches for file changes, so regex searches only read files that could match instead of scanning every file.
• openinterpreter/openinterpreter
  A terminal coding agent forked from OpenAI's Codex and tuned for low-cost open models such as Kimi K3. It can switch between emulated agent harnesses, such as claude-code or qwen-code, and works in ACP-compatible editors.
• pacifio/atlas
  Source control for coding agents. Each checkpoint links a commit to the session, prompts, tool calls and reasoning behind it, and agents such as Claude Code and Codex run side by side on one codebase with shared memory.

Only repositories in the ranked Trendshift lists are included, and the lists are used only to find candidates. We do not copy their ranks. Descriptions, licenses and star counts come from each GitHub repository. The notes are our own reading. We have not tested these projects, and a place on a trending list does not prove quality.