skills/jeremylongshore/claude-code-plugins-plus-skills/csrf-protection-validator/Gen Agent Trust Hub
csrf-protection-validator
Warn
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
- Command Execution (MEDIUM): The manifest explicitly allows 'Bash(npm:*)'. This permission allows the agent to execute any arbitrary npm package or command. For a security validation skill, this permission is excessively broad and could be exploited if the agent is prompted to run malicious code under the guise of validation.
- Indirect Prompt Injection (MEDIUM): The skill's purpose is to validate external security patterns. Because it possesses 'Write' and 'Bash' capabilities, there is a risk that malicious instructions embedded in the code being validated could trigger unintended side effects via these powerful tools.
Audit Metadata