secure-mobile-dev-guide
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input through the
$ARGUMENTSplaceholder to provide security guidance. While the primary purpose is informational, the agent environment allows high-privilege tools such asBashandTask. The instructions lack delimiters, boundary markers, or sanitization rules to prevent the agent from misinterpreting instructions embedded within user queries as directives for its tools. - Ingestion points: User queries are interpolated into the prompt via the
$ARGUMENTSvariable inSKILL.md. - Boundary markers: Absent. There are no delimiters (e.g., XML tags or triple quotes) or specific instructions for the agent to treat the input as pure data rather than potential instructions.
- Capability inventory: The skill is granted access to
Bash,Task,Read,Glob,Grep, andWebSearchtools. - Sanitization: Absent. No filtering, validation, or escaping of the user-provided content is performed before processing.
Audit Metadata