Agents In Use
What’s new todayLast update
18 additions

Security

These security AI agent use cases include accounts involving Claude, Codex. Each entry links its source and describes the task, reported outcome and human checkpoint.

Examples include “Claude Code adjusts a live Cloudflare search challenge” (Claude); “Codex builds a rate limiter for an over-crawled site” (Codex).

In “Claude Code adjusts a live Cloudflare search challenge”, Willison chose the rule behavior and created an expiring token with Zone WAF edit permission. In “Codex builds a rate limiter for an over-crawled site”, Willison selected the production configuration. The note does not document a separate approval or review protocol.

Security use cases

2 use cases

Frequently asked questions

What are real examples of AI agents for security?

Examples include “Claude Code adjusts a live Cloudflare search challenge” (Claude); “Codex builds a rate limiter for an over-crawled site” (Codex). Open each entry for the steps and reported outcome.

What human checkpoints appear in security agent use cases?

In “Claude Code adjusts a live Cloudflare search challenge”, Willison chose the rule behavior and created an expiring token with Zone WAF edit permission. In “Codex builds a rate limiter for an over-crawled site”, Willison selected the production configuration. The note does not document a separate approval or review protocol.

Which agents appear in security use cases?

The sourced entries in this category name Claude, Codex. Open each account to see the tools used and the author’s reported result.

Is it safe to delegate security tasks to an AI agent?

These accounts document particular runs rather than establish safety for every task. The recorded human checkpoints are: In “Claude Code adjusts a live Cloudflare search challenge”, Willison chose the rule behavior and created an expiring token with Zone WAF edit permission. In “Codex builds a rate limiter for an over-crawled site”, Willison selected the production configuration. The note does not document a separate approval or review protocol. Review the linked source and permissions before trying an adapted prompt.