The Agent Skills Directory

[SAFE]: The skill is a security-focused tool designed to scan and validate the safety of other AI agent skills. It contains no malicious patterns, unauthorized data access, or dangerous command execution.
[REMOTE_CODE_EXECUTION]: The skill includes TypeScript source code for its scanning logic, but it does not execute remote code, download external scripts, or use dangerous functions like eval() or exec().
[DATA_EXFILTRATION]: The skill defines regex patterns to detect data exfiltration attempts in other skills (e.g., searching for 'pastebin.com', 'webhook', or 'send all credentials') but does not perform any exfiltration itself. All network-related patterns are used for detection purposes.
[PROMPT_INJECTION]: No prompt injection or behavior override patterns were detected. The instructions are focused on providing a security role for the agent.
[COMMAND_EXECUTION]: No shell command execution or privilege escalation patterns were found. The embedded logic is limited to string processing, regex matching, and cryptographic hashing using standard libraries.

taskingbot-skill-validator