argent-react-native-app-workflow

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes project files and logs that could contain malicious instructions designed to influence the agent's behavior.\n
  • Ingestion points: 'package.json', 'metro.config.js', console log files, and native diagnostic reports.\n
  • Boundary markers: None present. The skill does not define delimiters or provide instructions to ignore embedded commands in the data being processed.\n
  • Capability inventory: Shell command execution (npm, yarn, npx), file system access, and JavaScript execution via 'debugger-evaluate'.\n
  • Sanitization: Content from external files is processed without explicit sanitization or validation.\n- [COMMAND_EXECUTION]: The skill uses shell commands to interact with the development environment, including 'lsof' to check ports and 'npm'/'yarn'/'npx' for build processes. These commands are often derived from the project's 'package.json'.\n- [DYNAMIC_EXECUTION]: The skill utilizes the 'debugger-evaluate' tool to execute arbitrary JavaScript code within the mobile application's runtime for debugging and state inspection purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 01:01 AM