x-navigation

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

• Third-party content exposure detected (high risk: 0.95). The workflow uses mcp__claude-in-chrome__read_page to extract visible tweets and profile text from X.com search results and user pages, which is outsider-authored free text from public web content fetched at runtime.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill explicitly navigates to and reads content from https://x.com at runtime and injects visible tweets/profiles into the agent context, meaning external content from that URL can directly influence prompts/behavior (flagged URL: https://x.com).