expo-gluestack-setup
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches UI component source code and configuration files from the official gluestack-ui GitHub repository (github.com/gluestack/gluestack-ui). This is the intended behavior for manual installation steps.\n- [COMMAND_EXECUTION]: Executes shell commands for package management (npm install, bun add) and tool initialization (npx gluestack-ui init). These operations are restricted to standard project tools and the official gluestack-ui CLI.\n- [REMOTE_CODE_EXECUTION]: Installs and runs packages from the NPM registry and executes initialization scripts via the gluestack-ui CLI. This behavior is expected as part of the skill's primary function to manage project dependencies.\n- [PROMPT_INJECTION]: The skill processes project configuration files and data from the GitHub API, representing a surface for indirect prompt injection.\n
- Ingestion points: Reads package.json, tailwind.config.js, metro.config.js, and GitHub API responses (references/v3.md).\n
- Boundary markers: No explicit delimiters are used for external data ingestion.\n
- Capability inventory: Has permissions to install packages, write to the filesystem, and execute CLI tools.\n
- Sanitization: Relies on verification of official documentation and source paths rather than programmatic sanitization.
Audit Metadata