The Agent Skills Directory

[SAFE]: The skill is documentation-heavy and primarily serves as a guide for security researchers. It does not bundle executable scripts or automate exploitation without user intervention.
[SAFE]: Includes a 'Mandatory confirmation gate' that requires the agent to verify the target URL, scope, and authorization before performing any probing or testing commands.
[COMMAND_EXECUTION]: Contains template curl and grep commands for manual security testing. These commands are intended to be executed within an authorized scope and use placeholders (e.g., YOUR_SESSION) for sensitive data.
[INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by processing user-provided target URLs for network operations. This risk is mitigated by the built-in requirement for explicit confirmation and the skill's primary purpose as a security testing tool.
[EXTERNAL_DOWNLOADS]: References official security disclosures and community repositories (GitHub, HackerOne) for educational context, following standard security research practices.

hunt-csrf