Hax — Development skill for Claude Code

Key features

• Lightweight by design — A single native C binary with a small dependency set. Starts instantly and uses only a few MB of memory, leaving more RAM for local models.
• Local models are first-class — Start llama-server -m [model].gguf, then hax --provider llama.cpp, and hax auto-discovers the model and runtime capabilities. No custom provider config block is needed for the default setup.
• Respects your terminal — Streaming Markdown and live tool output, reflowed for display in the terminal. Only redraws the current streaming line or the input area, native scrollback is preserved. Does not take over or mess with your terminal.
• Inspectable — See exactly what was sent to the model and what it replied in a usable transcript view (Ctrl+T). Optionally collect a detailed wire protocol trace.
• Broad provider support — OpenAI (+compatible), Anthropic (+compatible), Codex (via a ChatGPT subscription), OpenRouter, OpenCode Zen/Go, llama.cpp, and custom endpoints.
• Well-behaved Unix tool — XDG paths, clean stdout in -p one-shot mode with resume hints on stderr, plain-text config and session files, composition via subprocesses instead of plugins.

Target audience

Developers who live in the terminal, run local models, audit what their tools do, package software for distros, or run agents where resources are scarce. If you want MCP marketplaces, a plugin runtime, IDE panels, or per-command permission prompts, other agents build exactly that — hax deliberately doesn't, and [docs/philosophy.md](./docs/philosophy.md) explains each omission and the pattern that covers the need.

If "fancy new AI tech in an old-school minimalist package" sounds like your vibe, you might like this.

In