Claude Code can draw now. An experiment in reviewing from inside the agent
Two weeks ago Claude Code shipped something that is easy to miss in a changelog full of fixes. Version 2.1.287 added what the release notes call "Claude Mods: plugins may now modify deeper behavior". That undersells it. A mod is code that runs inside Claude Code, sees every tool call before it happens, and can draw its own interface in the terminal. I build Reado , an IDE for reading the code…
Claude Code has introduced a new feature called Claude Mods that allows plugins to modify its behavior and draw interfaces in the terminal. The author, who built an IDE for reading agent-written code called Reado, experimented with this feature. Claude Mods are JavaScript or TypeScript event handlers that run inside Claude Code's process, enabling it to draw its own interface in the terminal.
The author built a small mod to test its capabilities, which they referred to as "tasksFor". This mod monitors Reado tasks and appears above the prompt when a flagged line is edited by Claude Code. If the edit replaces the exact text mentioned in the comment, it highlights that change. The mod demonstrates the potential for moving reviewing an agent's work into the agent itself, with checks happening along the way instead of at the end.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
