The agent flow system for token maxxing.

Needs Python 3.12 or newer, uv, and a coding agent CLI you are signed in to, such as Claude Code or Codex.

uv tool install git+https://github.com/humanfia/humanize.git

Three backends need a package of their own: [dsh] for DeepSeek Harness, [kimi] for Kimi Code, [litellm] for a model called directly, [all] for all three.

uv tool install 'hmz[all] @ git+https://github.com/humanfia/humanize.git'

Open the terminal interface in a repository:

hmz

Or run a flow from your shell:

hmz exec -f chat -a assistant=claude/claude-opus-5:high "What does this repository do?"

Agents run with approvals bypassed, so start in a scratch repository. The documentation opens with a quickstart.

Pull requests are welcome: CONTRIBUTING.md says how to propose one. Ask a question, report a bug or request a feature through the issue forms, and report a vulnerability privately, as SECURITY.md says. Everyone taking part follows the Code of Conduct.

Apache-2.0 © Humanfia