twitter-downloader

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

• Third-party content exposure detected (high risk: 0.90). The helper script fetches tweet JSON from public third‑party endpoints (api.fxtwitter.com / api.vxtwitter.com) and the Agent workflow pastes the script stdout (tweet text, summary, and media links) directly into chat, meaning arbitrary user-generated tweet content is ingested and rendered where it could embed instructions that influence subsequent behavior.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.70). The skill instructs the agent to run a helper script that auto-installs missing packages using "apk add --no-cache" and writes files under /var, which entails modifying system packages/files and may require privileged (sudo) operations, so it pushes the agent to change the machine state.