mktg-x

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text from X (tweet/thread/bookmark/article/timeline body text) is fetched at runtime via the authenticated X GraphQL API using session cookies, then sanitized and inserted into the agent’s grounding memo as tweet_text/thread_unroll (untrusted third-party content that can contain indirect prompt-injection payloads).