The Agent Skills Directory

[SAFE]: The skill is a comprehensive reference guide for security audits, focusing on 'pit-of-success' design principles.\n- [COMMAND_EXECUTION]: Reference markdown files contain examples of shell command injection, such as rm -rf / in PHP and Python snippets. These were flagged by static detectors but are clearly labeled as insecure examples for auditing purposes.\n- [REMOTE_CODE_EXECUTION]: The documentation includes patterns for detecting dangerous dynamic execution, such as eval() and exec() in Ruby, Python, and JavaScript. These are provided as reference patterns and are not executed by the skill itself.\n- [SAFE]: The skill consists entirely of static documentation and lacks any executable scripts, network requests, or credential-gathering logic.

sharp-edges