smarty
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [EXTERNAL_DOWNLOADS]: Instructs the user to install the '@membranehq/cli' package from the official NPM registry, which is a trusted vendor resource for Membrane.
- [COMMAND_EXECUTION]: Employs the 'membrane' CLI for authentication and API communication, ensuring that credentials are managed securely by the platform.
- [INDIRECT_PROMPT_INJECTION]: The skill processes data returned from the Smarty API, creating a surface for potential instruction injection via external content.
- Ingestion points: Data payloads from 'membrane action run' and 'membrane request' outputs.
- Boundary markers: None identified in the provided instructions.
- Capability inventory: Execution of Membrane CLI commands and network proxying.
- Sanitization: No explicit sanitization of API response data is documented.
- [METADATA_POISONING]: The skill's description and text body contain significant inconsistencies, describing Smarty variously as a CRM, a PHP template engine, and an address verification service. This likely stems from documentation errors rather than malicious deception, as the infrastructure used (CLI, domains, and repositories) is consistently linked to the official vendor.
Audit Metadata