bane
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a structured prompt for performing systematic weakness analysis (pre-mortems) on user-provided content. It does not define any executable code or include external dependencies.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied plans, systems, or arguments, which could theoretically contain hidden instructions. However, the risk is mitigated by specific procedural steps and the restricted nature of the output.
- Ingestion points: The skill accepts a 'target' (plan, system, argument, or design) to analyze as described in
SKILL.md(Procedure Step 2). - Boundary markers: The skill explicitly instructs the agent to restate 'no-touch boundaries' before taking action (Procedure Step 1).
- Capability inventory: The skill uses only the agent's default toolset and does not request additional network or file-system permissions.
- Sanitization: No specific sanitization methods for input data are mentioned.
Audit Metadata