s4h-writing
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access detected. The skill consists of markdown instructions for agent behavior and tool routing.
- [PROMPT_INJECTION]: Indirect Prompt Injection surface: The skill accepts user-provided text passages which could contain instructions intended to influence downstream skills. 1. Ingestion points: Step 1 of SKILL.md (collecting user writing samples). 2. Boundary markers: Absent; user text is processed directly to diagnose problem clusters. 3. Capability inventory: The skill triggers other writing skills (e.g., /s4h-writing-character-development) based on the input. 4. Sanitization: No explicit filtering or sanitization of the user-provided text is performed before it is passed to the recommended skills as context.
Audit Metadata