creative-writer
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [No Code] (SAFE): The skill consists entirely of instructional markdown with no executable scripts, shell commands, or programming logic, which eliminates typical software-based attack vectors.
- [Prompt Injection] (SAFE): The instructions focus on creative formatting and literary styles. There are no attempts to bypass AI safety guards or extract system prompts.
- [Data Exposure & Exfiltration] (SAFE): No hardcoded secrets, sensitive file paths, or network communication tools (like curl or fetch) are present.
- [Indirect Prompt Injection] (SAFE): The skill has a low attack surface for indirect injection. Evidence: 1. Ingestion points: Processes user-provided 'theme or prompt' and 'constraints' in SKILL.md. 2. Boundary markers: None specified. 3. Capability inventory: No scripts, network access, or file-system permissions are defined for this skill. 4. Sanitization: None specified. Because there are no dangerous capabilities to exploit, the risk is negligible.
Audit Metadata