cubejs
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the official Membrane CLI (
@membranehq/cli) via the npm registry. This is a legitimate tool provided by the vendor for managing connections. - [INDIRECT_PROMPT_INJECTION]: The skill processes data from external Cube.js APIs, which is an inherent attack surface for indirect prompt injection.
- Ingestion points: API responses from Cube.js via
membrane action runormembrane requestcommands (SKILL.md). - Boundary markers: No specific instructions for using delimiters or ignore-markers are provided for the agent when processing external data.
- Capability inventory: Uses the
membraneCLI to perform network operations and data retrieval. - Sanitization: The instructions do not specify any explicit sanitization or validation steps for data returned from the Cube.js API.
Audit Metadata