security-osint

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill instructs the agent to ask the user for an API key or authentication code and embed it into code templates (e.g. client = XpozClient("THE_KEY_FROM_USER") or environment variables), passing secrets directly through the context and output.

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

• Third-party content exposure detected (medium risk: 0.30). The required workflow uses Xpoz to monitor and ingest social media feeds and posts from Twitter/X and Reddit, which are outsider-authored free text sources.