Skills
skills/reopt-ai/reopt-skills/opt-ui-install/Gen Agent Trust Hub

opt-ui-install

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from project configuration files listed in command/opt-ui-install.md (including package.json, .npmrc, and Tailwind configs) to detect project state. No explicit boundary markers or sanitization are specified. The skill's capability inventory includes shell-based package installation and file writing across the project structure, used here for legitimate integration fixes.
  • [PERSISTENCE_MECHANISMS]: The skill includes instructions to persist the GITHUB_TOKEN environment variable by appending an export command to shell profile files (~/.zshrc, ~/.bashrc) as described in references/fix-config.md.
  • [DYNAMIC_EXECUTION]: The workflow in command/opt-ui-install.md generates a theme boot script using createThemeBootScript() and injects it into the root layout to handle client-side theme initialization.
  • [COMMAND_EXECUTION]: The skill executes shell commands using npm, bun, and npx to manage dependencies and interact with the @reopt-ai/opt-ui-cli tool.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 07:23 AM