flutter-mobile-automation

Pass

Audited by Gen Agent Trust Hub on Oct 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses standard command-line tools for mobile development, including flutter, dart, adb, maestro, and xcrun. These are employed for building, installing, and testing Flutter applications as part of a development and QA workflow.
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches resources from github.com and docs.maestro.dev. Specifically, it directs the agent to the official Dart team repository (dart-lang/ai) and the documentation for Maestro. These are well-known and reputable sources in the development ecosystem.
  • [INDIRECT_PROMPT_INJECTION]: The skill documents an attack surface where the agent processes data from external sources during execution.
  • Ingestion points: The agent is instructed to read application logs via get_app_logs and the Flutter internal structure via get_widget_tree in SKILL.md.
  • Boundary markers: The skill does not explicitly define delimiters or instructions to ignore embedded commands within the ingested logs or widget data.
  • Capability inventory: The agent has access to shell execution via tools like flutter, dart, and maestro, as defined in the permissions section of SKILL.md.
  • Sanitization: No specific sanitization or filtering logic is provided for the data ingested from the running application.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 3, 2026, 11:25 PM