Back to directory

ZAI Shell

An autonomous terminal agent that turns natural-language intent into verified system actions on Windows, Linux and Mac, with an --auto mode that executes, checks its own output and retries until a task completes.

ZAI Shell

ZAI Shell translates natural-language requests into system commands and, in its --auto mode, executes each step, reads the resulting terminal output, and adjusts or retries until the task is complete rather than stopping after one command. Several operating modes trade reasoning depth for speed or lower API usage: a repair-focused fixer mode, a fast lightning mode, a quota-conscious eco mode, and a balanced normal mode.

A "Sentinel" component warns before running a command it judges destructive and can block prompt-injection attempts found while reading files; every file edit is backed up automatically, and a single undo command restores the most recent one. A --watch flag can monitor a condition such as CPU usage and alert when it is met.

The agent keeps the input and output of its last five actions as working memory, and it can take a screenshot on a hotkey to read an on-screen error directly instead of requiring it to be copied and pasted in.

More in Coding agents

See all 793
  • The City

    github.com

    A native macOS app that shows Claude Code runs as a city, with each project as a building, each team as a floor of robots, and every animation driven by a real event from the claude CLI's stream.

  • rig

    tryrig.ai

    A small, single-binary agent runtime that points at any OpenAI-compatible model, local or hosted, reads and edits a repository's code, runs its tests and remembers what it learned, with a loop implemented in a few hundred lines.

  • Jauvex

    jauvex.reindent.com

    An Electron desktop app that puts Claude Code, Codex and Grok coding-agent sessions side by side and lets you steer them by voice, with a small model handling fast routing decisions.