android-review
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to review untrusted Android source code and diffs, which serves as a potential vector for instructions embedded within the code targeting the agent.
- Ingestion points: Android source code, manifests, and diffs provided for analysis (referenced in SKILL.md and current-guidance.md).
- Boundary markers: The instructions do not explicitly define boundary markers to isolate the analyzed code from the agent's core instructions.
- Capability inventory: The skill utilizes static analysis tools (ktlint, detekt), testing frameworks (JUnit, Espresso), and interacting with devices via the device MCP (Model Context Protocol).
- Sanitization: No specific input sanitization or escaping of the ingested code is described, although the skill's primary focus is the detection of security risks within that data.
- Risk Assessment: This surface is inherent to the intended primary purpose of a code review tool and is considered safe in this context, especially given the presence of defensive checks for security-sensitive surfaces.
Audit Metadata