tweet-ideas

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

• Third-party content exposure detected (high risk: 1.00). The skill's SKILL.md includes a GET /trends endpoint and a "seed_from_trends" option (via https://xquik.com/api/v1) that ingests public X/Twitter trending content (user-generated) to seed and influence generated tweet ideas, so third-party content is read and can change agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill calls the runtime API at https://xquik.com/api/v1 (e.g., POST /compose, GET /trends) to fetch "ideas" whose returned prompt text is then used to drive subsequent agent instructions, so external content fetched at runtime directly controls prompts.