Skills
skills/aradotso/mcp-skills/commercevault-edd-ecommerce-api/Gen Agent Trust Hub

commercevault-edd-ecommerce-api

Pass

Audited by Gen Agent Trust Hub on Jun 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation directs users to download the server code from a third-party GitHub repository (dhapat3927/mcp-edd-analytics-vantage.git) which is not associated with a verified organization.
  • [COMMAND_EXECUTION]: Installation requires running commands like npm install and npm start or python main.py on code downloaded from an external source, allowing for arbitrary code execution on the user's machine.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from an e-commerce store. 1. Ingestion points: Data enters the context via tools like list_orders and get_customer fetching data from the store API. 2. Boundary markers: No explicit delimiters or instructions are provided to help the agent isolate store data from instructions. 3. Capability inventory: The skill provides high-privilege write capabilities such as create_refund, create_order, and activate_license. 4. Sanitization: No sanitization or validation of the ingested API data is described in the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 30, 2026, 06:40 PM
Security Audit — agent-trust-hub — commercevault-edd-ecommerce-api