oma-design
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches design templates and manifests from the npm registry and the getdesign catalog.
- Evidence:
resources/getdesign-fetcher.mddescribes the process of fetching the latestgetdesignmanifest and templates fromhttps://registry.npmjs.org/getdesign/latest. - Security Control: The skill implements mandatory SHA256 integrity verification (hash pinning) for all downloaded templates before they are processed by the agent.
- [COMMAND_EXECUTION]: The skill invokes various CLI tools via
bunxandnpxfor design template management and component installation. - Evidence:
SKILL.mdandresources/getdesign-fetcher.mdutilizebunx getdesign@latestfor fetching assets.resources/component-patterns.mdrecommendsnpx shadcn@latestfor component management. - [PROMPT_INJECTION]: The skill identifies the risk of indirect prompt injection from external design data and provides explicit defensive instructions.
- Evidence:
resources/getdesign-fetcher.mdcontains a 'Prompt-Injection Defense' section that instructs the agent to treat external data as 'REFERENCE DATA ONLY' and to ignore any imperative instructions or meta-instructions within external files.
Audit Metadata