cloud-storage-web
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from external storage, which represents a potential attack surface for indirect prompt injection.
- Ingestion points: The skill describes tools for downloading and previewing files via
app.downloadFileandapp.getTempFileURLinSKILL.md. - Boundary markers: No explicit markers or instructions to delimit or ignore embedded commands are present in the provided instructions.
- Capability inventory: The skill enables file manipulation (upload and deletion) and database interactions through
managePgDatabaseas specified inSKILL.md. - Sanitization: The instructions do not define specific content validation or filtering for data retrieved from storage.
- [DYNAMIC_EXECUTION]: The skill provides SQL templates to be executed via the
managePgDatabasetool to configure Row Level Security (RLS) on storage tables. This is a legitimate administrative task designed to implement least-privilege access controls within the environment.
Audit Metadata