x

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

• Third-party content exposure detected (medium risk: 0.30). The skill reads X (Twitter) feeds, timelines, and search results containing outsider-authored free text without the agent first selecting a specific item.

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

• Potentially malicious external URL detected (high risk: 0.70). The script dynamically fetches a client-web chunk script from abs.twimg.com at runtime to resolve transaction IDs for Twitter's internal web API, constituting a weak/unknown runtime operational dependency.