signals-scout-skills-store

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads skills from the PostHog skills store via tool calls (skill-list, skill-get, skill-file-get), which contain outsider-authored text (skill names, descriptions, bodies, and files) without prior selection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (medium risk: 0.30). The URLs in line 95 are static documentation and specification references instructing the agent to refresh its ruleset, constituting an optional/fallback operational dependency on third-party documentation resources.