flutter-device-crud-testing
Pass
Audited by Gen Agent Trust Hub on Jul 1, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides specific shell commands for process management and troubleshooting in Phase 1.5 and Phase 5.13. These include the use of
ps auxandpkillto terminate lingering development processes such asflutter_tools,dart.*tooling_daemon, andGradleDaemon. While standard for local development environments, these represent executable commands provided to the agent. - [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it requires the agent to ingest and act upon data from external, potentially untrusted sources.
- Ingestion points: The agent is instructed to read project source files (router and screen definitions in Phase 0) and parse UI element metadata (text labels and accessibility tags in Phase 2, 5, and 6) to determine its next actions.
- Boundary markers: Absent. There are no instructions or delimiters provided to help the agent distinguish between its own system instructions and text content found within the app's UI or source code.
- Capability inventory: The skill possesses significant capabilities, including UI interaction (tapping, typing, swiping), app lifecycle control (launching and reloading apps), and shell command execution (
pkill). - Sanitization: Absent. The instructions do not include validation or filtering steps for the data parsed from the application environment.
Audit Metadata