hormuz-risk-clock

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.85). The skill mandates automated fetching of fresh web signals and automatic publication of rendered outputs to hardcoded external accounts/channels (Discord channel ID and Bluesky via env vars) without requiring user approval, creating a high risk of unauthorized data exfiltration or covert publication.

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

• Third-party content exposure detected (high risk: 0.95). Runtime path: the workflow explicitly uses websearch + webfetch to pull full text from fresh public web sources, and that fetched outsider-authored prose is then normalized into data/signals.latest.json and folded into the LLM’s context for state update/render.