mobile-security
Pass
Audited by Gen Agent Trust Hub on Apr 15, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection when analyzing untrusted mobile application source code. This occurs because the skill ingests external project data but lacks instructions to the agent to treat that data as non-authoritative.
- Ingestion points: The skill searches for and reads various project files including AndroidManifest.xml, build.gradle, Info.plist, package.json, and source code files (*.kt, *.swift, *.dart).
- Boundary markers: No explicit delimiters or isolation instructions are present to prevent natural language commands within the analyzed code from influencing the agent's execution.
- Capability inventory: The skill performs pattern matching and code inspection on untrusted file content.
- Sanitization: No sanitization or filtering of analyzed content is specified.
Audit Metadata