A shared runtime policy library for Claude Code

[ishita-0301]

One thing I'd love to see emerge around Claude Code is a community-maintained collection of runtime policies, not just prompts or hooks.

For example:

• Stop execution if the same tool is called 5 times with no progress.
• Pause before destructive file operations.
• Require confirmation before force pushes or mass deletions.
• Detect when an agent is retrying the same failing command over and over.
• Surface a warning when token usage spikes unexpectedly.

These feel like reusable building blocks that many developers would want, regardless of the project they're working on.

I recently came across FailproofAI (https://github.com/FailproofAI/failproofai), which explores some of these ideas. It made me wonder if the Claude Code ecosystem could benefit from a standardized library of execution guardrails in the same way we share skills and hooks today.

Curious if others have built similar policies or think this would be useful.