mobile-security
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface (Category 8). The skill is designed to ingest and analyze external data (source code and project configuration files), which presents a potential surface for indirect instructions.
- Ingestion points: Mobile project files including AndroidManifest.xml, build.gradle, Info.plist, package.json, pubspec.yaml, and source code files (*.kt, *.java, *.swift, *.dart, *.tsx).
- Boundary markers: Absent. The skill does not explicitly define delimiters to separate user data from its instructions or warn the agent to ignore embedded instructions in the audited files.
- Capability inventory: The skill is a pure analysis tool; it does not request tool permissions or shell execution capabilities.
- Sanitization: Absent. The skill performs direct pattern matching on the file contents.
- [SAFE]: The skill includes example credentials and URLs (e.g., 'sk-live-a1b2c3d4e5f6g7h8i9j0', 'api.example.com') for illustrative purposes within its 'UNSAFE' code snippets. These are benign placeholders used to teach the agent which patterns to detect during an audit and do not represent actual secrets or exfiltration endpoints.
Audit Metadata