skills/vm0-ai/vm0-skills/zendesk/Gen Agent Trust Hub

zendesk

Pass

Audited by Gen Agent Trust Hub on Apr 17, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE]: The skill utilizes environment variables for authentication, instructing the user to store sensitive information like ZENDESK_API_TOKEN and ZENDESK_EMAIL securely rather than hardcoding them in the instructions or scripts.
  • [EXTERNAL_DOWNLOADS]: All network operations are directed towards well-known and official domains, specifically *.zendesk.com and developer.zendesk.com, for API interaction and documentation.
  • [COMMAND_EXECUTION]: Uses common command-line utilities including curl for REST API communication, jq for processing JSON responses, and sed for local file manipulation of request payloads.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection because it retrieves and processes untrusted data from Zendesk tickets (such as subject lines and comment bodies). It lacks explicit boundary delimiters to separate this external content from the agent's primary instructions, though this is inherent to its purpose of managing support tickets.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 17, 2026, 04:44 PM