duck-simplify
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: Static analysis of the skill's instructions and referenced guardrails confirms it has no malicious functionality. It does not perform network requests, modify the file system, or execute shell commands. The behavior is limited to generating text-based review findings.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted code and proposals for review, which constitutes an indirect prompt injection surface. However, the risk is mitigated by the skill's lack of executable capabilities and its safety-focused instructions.\n
- Ingestion points: Code and proposals provided for simplification review in the Method section (SKILL.md).\n
- Boundary markers: Not explicitly defined for input ingestion, although the skill uses a structured output format with specific prefixes.\n
- Capability inventory: The skill is restricted to text generation and does not have access to tools for system modification, network communication, or code execution.\n
- Sanitization: Safety carve-outs in the shared guardrails (references/GUARDRAILS.md) explicitly direct the agent to refuse any suggestions that would weaken security controls, trust boundaries, or accessibility requirements.
Audit Metadata