knock-cli

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

• Third-party content exposure detected (high risk: 0.90). This skill's Workflow templates (see the "HTTP fetch steps" section in SKILL.md / rules/workflow-templates.md) documents an http_fetch step that issues HTTP requests to arbitrary URLs (e.g., "url": "https://api.example.com/..."), shallow-merges the JSON response into trigger data, and uses that data in subsequent steps—demonstrating ingestion of untrusted third-party content that can materially influence workflow behavior.