design-pattern-advisor
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill operates by ingesting and analyzing external content, including source code, modules, and architecture proposals. This creates a surface for indirect prompt injection where instructions embedded in analyzed data could potentially influence the agent's behavior.\n
- Ingestion points: The skill reads code, pull requests, and repository artifacts as specified in the 'Required context' section of
SKILL.md.\n - Boundary markers: The instructions focus the agent on specific design problems but do not explicitly define boundary delimiters or provide 'ignore instructions' warnings for the external content being processed.\n
- Capability inventory: The skill advises using tools such as ripgrep, AST tools, ESLint, and various static analysis utilities to gather evidence from the filesystem and GitHub.\n
- Sanitization: The skill instructions do not specify any sanitization or escaping protocols for the content ingested during analysis.\n- [SAFE]: The skill includes a local Python script
scripts/validate_scenarios.pywhich is used for static contract validation of the skill's own documentation and structure. This script performs simple string matches on local files and does not engage in network operations, dynamic code execution of untrusted input, or sensitive data access.
Audit Metadata