seven-prompt-content-engine
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill operates as a purely instructional framework for content generation and platform adaptation. Analysis of the instructions and prompt templates revealed no malicious intent, obfuscation, or unauthorized access attempts.\n- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest user-provided ideas and content from external URLs to use as source material for drafting.\n
- Ingestion points: User-supplied rough ideas, writing samples, and external source material URLs (SKILL.md, intake section).\n
- Boundary markers: The prompt templates use bracketed placeholders (e.g., [ROUGH_IDEA]) but lack explicit instructions or delimiters to isolate untrusted data from the core instructions.\n
- Capability inventory: The skill performs text generation and platform-native formatting. It does not manifest dangerous capabilities such as shell execution, file system modification, or network exfiltration of sensitive data.\n
- Sanitization: Step 5 (Humanizer) includes an audit pass that explicitly checks for fake specificity or invented numbers not grounded in the user's material, providing a verification layer against unauthorized content injection.
Audit Metadata