x-bookmarks-fetcher

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

• Third-party content exposure detected (high risk: 0.75). The skill calls the X API at runtime and ingests outsider-authored tweet content (e.g., t.note_tweet?.text / t.text) from GET /2/users/:id/bookmarks into local JSON/MD/TXT files, which are readable prose and could be fed into an agent LLM context indirectly via those outputs.