zendesk

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt includes non-interactive examples and commands that pass API tokens, client secrets, and webhook/webhook URLs directly as command-line flags and in examples, which would require an agent to insert secret values verbatim into generated commands or code.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly fetches and saves user-generated Zendesk content (e.g., via ticket-details, search, saved JSON files and uv run zd-cli query described in SKILL.md) and can download arbitrary attachment URLs (attachment command), and those saved comments/attachments are parsed by analysis scripts (e.g., analyze_support_metrics.py, call_mentions queries) and used to drive reports and potential write actions (create-ticket, add-comment), so untrusted third‑party content from Zendesk/attachment URLs is ingested and can materially influence subsequent tool actions.