agent-add-rule
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns or security risks were detected. The skill's functionality is focused on documentation management and organizing agent instructions within a project.
- [PROMPT_INJECTION]: The skill accepts user-supplied instructions for storage in high-context configuration files like CLAUDE.md, creating an indirect prompt injection surface. Ingestion points: Free-text user input for new rules and existing documentation files. Boundary markers: No explicit delimiters or 'ignore' markers are used for user-supplied content. Capability inventory: The skill is granted Write, Edit, and Bash tools to modify project files. Sanitization: No validation or sanitization is performed on the rule content before it is written to the filesystem.
Audit Metadata