usmds
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches a component registry from a remote URL to identify available components.
- Evidence: Programmatically retrieves JSON data from
https://storage.googleapis.com/usmds-registry/r/usa/registry.jsoninreferences/components.md. - [COMMAND_EXECUTION]: Documentation suggests executing a vendor-provided initialization tool.
- Evidence: Recommends running
npx @blen/usmds initfor automated project setup inSKILL.mdandreferences/setup.md. - [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from a remote registry which informs agent output.
- Ingestion points: Reads external JSON registry from
https://storage.googleapis.com/usmds-registry/r/usa/registry.jsoninreferences/components.md. - Boundary markers: None identified; the registry data is integrated into the agent's knowledge of available components.
- Capability inventory: The skill uses registry items to generate React Native code and documentation.
- Sanitization: No validation or sanitization of the remote JSON content is documented.
Audit Metadata