skills/mikefilsaime-groove/clickcampaigns-for-claude-code-in-cursor/book-writing/Gen Agent Trust Hub
book-writing
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [PROMPT_INJECTION] (SAFE): No evidence of instructions designed to bypass safety filters or extract system prompts. The instructions are focused solely on book writing structure and style.
- [DATA_EXFILTRATION] (SAFE): No network operations or sensitive file path references detected.
- [REMOTE_CODE_EXECUTION] (SAFE): No external downloads, package installations, or shell command executions found.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill is designed to process user-provided book concepts. While this is an ingestion point for untrusted data, the skill only produces text output and lacks capabilities (like file writing or network access) that could be exploited via injection.
Audit Metadata