social-publisher

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

• Third-party content exposure detected (medium risk: 0.30). The workflow optionally uses TweetClaw to search public tweets and replies, which exposes the agent to outsider-authored free text from social media feeds.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs setup and key validation via an unverified third-party domain (getsocialclaw.com) and an external npm plugin installation (npm:@xquik/tweetclaw@1.6.31), constituting weak/unknown operational dependencies.