relational-database-web-cloudbase

Pass

Audited by Gen Agent Trust Hub on Nov 12, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill facilitates the retrieval of data from a relational database (MySQL) using the CloudBase SDK. This creates an indirect prompt injection surface where maliciously crafted content stored in the database could attempt to override agent instructions when the results are processed.
  • Ingestion points: Results from database queries performed via db.from().select() (SKILL.md).
  • Boundary markers: The provided examples do not include delimiters or instructions to ignore embedded commands in the retrieved data.
  • Capability inventory: The skill enables data retrieval, insertion, and deletion operations via the @cloudbase/js-sdk.
  • Sanitization: No specific output sanitization or validation logic is demonstrated in the implementation examples.
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @cloudbase/js-sdk package via npm. This is a standard vendor-provided library necessary for the skill's stated purpose of interacting with CloudBase services.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 12, 2026, 02:29 AM