expo-gluestack-setup

Pass

Audited by Gen Agent Trust Hub on Oct 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses official package managers (npx and bunx) to download and execute the gluestack-ui CLI tool. It also references official documentation from gluestack.io and source code from github.com/gluestack, which are established and well-known repositories in the mobile development ecosystem.\n- [COMMAND_EXECUTION]: The skill runs diagnostic commands such as gluestack-ui init --help to confirm environment capabilities. It further generates precise shell commands for the user to execute manually in their terminal, maintaining a 'human-in-the-loop' security model for project modifications.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes local project files to determine the current state of the application and apply necessary configuration updates.\n
  • Ingestion points: Project files including package.json, tailwind.config.js, metro.config.js, and global.css are inspected to verify prerequisites and setup status.\n
  • Boundary markers: The skill follows a strict Gluestack Handoff format and outcome tracking to ensure the agent maintains clear boundaries between its analysis and project modification steps.\n
  • Capability inventory: The skill possesses the ability to read and update project configuration files and execute non-interactive CLI help commands.\n
  • Sanitization: Verification criteria are established to validate that generated provider and component exports resolve correctly and that configuration rewrites remain consistent with official framework requirements.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 7, 2026, 04:43 AM