web-design-guidelines
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [Indirect Prompt Injection Surface]: The skill fetches external compliance instructions dynamically from a remote repository, which creates an inherent surface for indirect prompt injection if the external resource is modified.
- Ingestion points: External instructions are retrieved via a URL specified in
SKILL.md. - Boundary markers: There are no explicit delimiters or boundary markers instructing the agent to isolate the fetched rules from its core instructions.
- Capability inventory: The skill is restricted to reading local code files and performing text analysis based on the retrieved guidelines, with no administrative or shell command execution capabilities.
- Sanitization: No data validation or validation schemas are applied to the fetched remote content.
- [External Resource Dependency]: The skill dynamically fetches configuration data from a remote URL at runtime, creating a dependency on external network access and the continuous integrity of the host repository.
Audit Metadata