android-app-submission

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions to execute various shell commands for generating release keystores, building the Android App Bundle (AAB), and verifying signatures.
  • Evidence: Commands such as keytool -genkeypair, ./gradlew :app:bundleRelease, and jarsigner -verify are listed in the prerequisites section of SKILL.md.
  • [DYNAMIC_EXECUTION]: The skill utilizes JavaScript execution through a browser automation tool to interact with the Google Play Console user interface, particularly to handle elements that are difficult to interact with via standard clicks.
  • Evidence: The instructions explicitly use browser_evaluate() to perform clicks on buttons identified by debug-id attributes in SKILL.md.
  • [INDIRECT_PROMPT_INJECTION]: The workflow relies on the content of a local CHANGELOG.md file to provide the text for Play Store release notes, creating a surface for potential injection if the file is modified by an untrusted source.
  • Ingestion points: The agent is instructed to read the CHANGELOG.md file at the repository root as the source of truth for the Play release notes.
  • Boundary markers: There are no specified boundary markers or instructions for the agent to ignore potentially malicious content within the changelog text.
  • Capability inventory: The skill uses Playwright MCP tools like browser_type and browser_evaluate to input the changelog data into the Google Play Console.
  • Sanitization: No validation or sanitization steps are described for the data extracted from the CHANGELOG.md file before it is used in automation steps.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 07:29 AM
Security Audit — agent-trust-hub — android-app-submission