gluestack-ui

Pass

Audited by Gen Agent Trust Hub on Sep 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses npx gluestack-ui@latest to download and add UI components from the public npm registry, which is a common practice for this well-known UI library.
  • [REMOTE_CODE_EXECUTION]: Execution of remote code is performed via npx for both the gluestack-ui CLI and the TypeScript compiler (tsc) to manage components and verify type safety.
  • [COMMAND_EXECUTION]: The instructions utilize local shell commands including grep, sort, and tsc to analyze source files, extract configuration tokens, and perform build-time checks.
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes a surface for indirect prompt injection by instructing the agent to ingest and act upon the contents of component source files to automate styling integrations.
  • Ingestion points: UI component source files located in src/components/<component>/index.tsx.
  • Boundary markers: No delimiters or explicit warnings to ignore instructions embedded in the source files are present.
  • Capability inventory: The skill utilizes shell command execution (via grep and npx) and has the capability to modify global style and theme files (global.css, theme.css).
  • Sanitization: Styling tokens extracted from source files are used to bind to theme variables without validation or escaping of the ingested content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 28, 2026, 04:07 PM