Anthropic Finally Let You Hack Claude Code — Expect More Doom Clones
Anthropic just dropped Claude Mods, letting devs use TypeScript to inject custom chaos into their AI coding assistant. Because clearly, what your professional IDE was missing was a built-in game of Doom.
Anthropic has opened early access to Claude Mods, essentially allowing developers to build plugins for Claude Code using simple TypeScript hooks. These mods intercept tool calls, manipulate data, and even force their way into the UI with custom buttons and panels.
The core functionality allows for real-time monitoring and interface injection. A basic example lets you measure exactly how long a Bash command takes to run, displaying a toast notification with the latency. More practical, though arguably less fun, examples include hiding sensitive data until a user hovers over it or tracking pull request statuses directly inside the console.
Getting started requires setting a specific environmental flag, CLAUDE_CODE_ENABLE_FUNCTION_HOOKS=1, to unlock the API. Naturally, the developer community has already ignored the 'professional' potential to focus on what matters: porting Tetris, breathing exercises, and the inevitable version of Doom directly above the command prompt.
It is truly heartwarming to watch humanity take a sophisticated Large Language Model designed for complex software architecture and immediately turn it into a dedicated gaming console. Surely, this is the pinnacle of engineering productivity everyone promised.
Source: sourcecraft.dev
Comments
This is where the magic happens: AI reads your discussion and rewrites the article based on the most interesting comments. Each strong comment adds points to the meter below. Once the meter is full, the article updates live — no page reload needed.