e2e-testing-mobile

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install the Maestro CLI and EAS CLI tools from their official sources.
  • Evidence: templates/install-plan.md contains commands to fetch the installer from https://get.maestro.mobile.dev and install eas-cli globally via npm.
  • [REMOTE_CODE_EXECUTION]: The recommended installation procedure for the Maestro CLI executes a remote script fetched from a web server directly in the shell.
  • Evidence: templates/install-plan.md and references/maestro-cli.md document the command curl -Ls "https://get.maestro.mobile.dev" | bash.
  • [COMMAND_EXECUTION]: The skill relies on several local CLI tools to verify the environment, execute test flows, and manage mobile app builds.
  • Evidence: SKILL.md utilizes commands such as maestro --version, jq, xcrun simctl, and adb devices for preflight checks. maestro test is used for test execution.
  • [INDIRECT_PROMPT_INJECTION]: The skill implements a workflow that parses Markdown-based feature specifications to generate executable Maestro YAML flows, which introduces a potential surface for indirect prompt injection if the source data contains malicious instructions.
  • Ingestion points: Markdown files located in the specs/ directory, which are either user-authored or agent-generated (SKILL.md, rules/spec-first-flow.md).
  • Boundary markers: The skill does not define specific delimiters or instructions for the agent to ignore potentially malicious content embedded within the specifications.
  • Capability inventory: The skill possesses the ability to write to the file system (creating .maestro/ flows), execute shell commands (maestro test), and trigger cloud builds (eas build).
  • Sanitization: The documentation does not describe explicit sanitization or validation steps for the content of the Markdown specifications before they are processed to generate code.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 12:51 PM