dotnet-design-pattern-review
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code provided by the user via the
${selection}variable. - Ingestion points: The
${selection}placeholder inSKILL.mdis the primary entry point for external data. - Boundary markers: No explicit delimiters or instructions are provided to the agent to help distinguish between the skill's instructions and potentially malicious instructions embedded within the code being reviewed.
- Capability inventory: While the skill's instructions explicitly forbid code modifications ("Do not make any changes to the code, just provide a review"), the YAML frontmatter does not utilize the
allowed-toolsfield to restrict the agent's default toolset, which may include file access or command execution depending on the host environment. - Sanitization: There is no evidence of input validation or sanitization for the content of the
${selection}variable.
Audit Metadata