hig-components-menus
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions require the agent to ingest project-specific context from a file within the workspace. This behavior introduces a surface where untrusted data could influence or override agent behavior.
- Ingestion points: .claude/apple-design-context.md (referenced in SKILL.md).
- Boundary markers: The skill does not define delimiters or provide instructions to ignore embedded commands within the context file.
- Capability inventory: While the skill is informational, the agent using it typically has access to terminal commands, file system operations, and network tools.
- Sanitization: No evidence of validation, sanitization, or filtering of the content read from the context file was found.
Audit Metadata