dev
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- Command Execution: The skill uses the
Bashtool to execute themsCLI for managing app development and employs system utilities (open,xdg-open,Start-Process) to launch the user's browser. These actions are standard for local development workflows and facilitate the skill's primary purpose. - Indirect Prompt Injection Surface: The skill monitors the output of the development server to extract an App Player URL, which is then passed to a shell command for opening. This creates a potential data-to-command flow that warrants review if the development server output can be influenced by untrusted external data.
- Ingestion points: The skill reads
memory-bank.mdfor project context and monitors the stdout of thems app devcommand (specifically looking for the "Ready" signal). - Boundary markers: The skill uses a specific string pattern ("Ready. You can play your app locally at: ") to identify the data to be captured.
- Capability inventory: The agent has access to the
Bashtool, which is used to run the CLI and system browser commands. - Sanitization: The instructions focus on capturing the exact URL output by the CLI without specifying additional escaping or validation before passing it to the browser-open command.
Audit Metadata