android-app-factory
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data including repository code, Play Console status, and official Google policy documentation, which represents an attack surface for indirect instructions. \n
- Ingestion points: The skill instructs the agent to read repository source code, live Play Console state, and external policy URLs (specified in
SKILL.mdandreferences/play-policy-baseline.md).\n - Boundary markers: The skill implements explicit 'Confirmation gates' for irreversible actions like Play Store uploads or track promotions (defined in
SKILL.md, Boundaries section), and requires the agent to document the source and time of policy checks.\n - Capability inventory: The agent is authorized to execute build and inspection tools such as
./gradlew,bundletool, andapkanalyzer(references/android-factory-pipeline.md), and can interact with the Google Play API for deployments.\n - Sanitization: There are no specific instructions for sanitizing or escaping content ingested from repository files or external web pages before processing.
Audit Metadata