solo-android-release
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [DATA_EXFILTRATION]: The skill instructs the agent to send telemetry data to an external endpoint (https://analytics.superduperai.co/e). While the data consists of anonymized identifiers (UUIDs) for install tracking, this represents a network operation to a non-whitelisted domain.- [INDIRECT_PROMPT_INJECTION]: The skill uses Playwright automation to interact with the Google Play Console, which involves reading untrusted content from web pages that could potentially influence agent behavior.
- Ingestion points: Reads document.body.innerText from the Google Play Console (play.google.com) during automation steps.
- Boundary markers: None identified; instructions do not specify delimiters or warnings for the agent when processing page content.
- Capability inventory: The skill utilizes browser_click, browser_evaluate, and browser_file_upload for browser automation and executes shell commands like gradlew and gh for build and release tasks.
- Sanitization: No sanitization or validation of the ingested web content is described before it is processed by the agent.- [COMMAND_EXECUTION]: The skill performs shell operations using gradlew, gh (GitHub CLI), and keytool. These are standard tools for building, signing, and releasing Android applications.
Audit Metadata