Anthropic introduces mods for Claude Code to customize behavior and UI

Anthropic has introduced mods, small TypeScript functions that change how Claude Code works. These mods can rewrite prompts, add new UI, replace built-in features, or add entirely new functionality. They ship inside plugins and are available today in the Claude Code CLI and desktop app.

How mods work

Each time Claude Code performs an action, it emits an event. A mod is a function that hooks into one of these events, running before, after, or instead of the event. With a single function, a mod can:

• Rewrite a prompt before it reaches the model.
• Block, rewrite, or retry a tool call.
• Approve or deny a permission request.
• Redact secrets from tool output.

Mods can also modify the interface, such as editing tool results or adding buttons. When multiple mods hook the same event, they run in the order they load, allowing users to stack mods from different authors.

Security and enterprise controls

Mods run with the same access to the machine as Claude Code itself and are not sandboxed. Users should only install mods from trusted sources. For Team and Enterprise plans, a built-in mod called sec-default loads first to prevent risky actions, such as overriding permission deny rules. Admins can manage plugin marketplaces and push managed settings to users' machines.

Teams can use mods to build custom controls, such as displaying CI/CD status, requiring confirmation for production commands, or implementing audit logging. Some built-in features, like /diff, are now mods themselves, allowing users to disable or replace them.