shaman-linus
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill employs strong persona-driven instructions to establish a specific character for code review. While the language used is confrontational (e.g., 'garbage code', 'you don't deserve to talk to me'), these instructions are intended for role-playing and do not attempt to bypass safety filters or extract system prompts.\n- [PROMPT_INJECTION]: The skill's function as a code analyzer creates a surface for indirect prompt injection from untrusted user code.\n
- Ingestion points: User-provided code blocks for review as specified in the 'Final Challenge' and 'Methodology' sections of SKILL.md and SKILL.v2.md.\n
- Boundary markers: Absent. No specific delimiters or instructions to ignore embedded commands within the analyzed code are provided.\n
- Capability inventory: The skill does not specify or ship with any subprocess, file-write, or network operations.\n
- Sanitization: Absent. The skill does not instruct the agent to sanitize or escape user-provided code before analysis.
Audit Metadata