mendix
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to install the
@membranehq/clipackage from the official NPM registry. This is the vendor-provided tool required for managing connections and interacting with the Membrane platform.\n- [COMMAND_EXECUTION]: Several CLI commands are provided for agent use, includingmembrane login,membrane connection ensure, andmembrane action run. These operations are essential for the integration's core functionality and facilitate secure communication with external APIs.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes data returned from Mendix actions and API requests, creating a vulnerability surface where external content could influence agent behavior.\n - Ingestion points: Data returned from action results (
membrane action run) and proxy requests (membrane request).\n - Boundary markers: The skill does not define specific delimiters or instructions to treat external data as untrusted or to ignore embedded instructions.\n
- Capability inventory: The skill enables the agent to execute actions and perform network requests through the Membrane infrastructure.\n
- Sanitization: No explicit logic for sanitizing, validating, or escaping external content is included in the instructions.
Audit Metadata