bibi-library

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

• Third-party content exposure detected (medium risk: 0.30). The agent skill requires browsing, searching, and managing saved videos, notes, and collections within a BibiGPT library, which ingests user-saved content and notes that may contain untrusted outsider-authored text.

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

• Potentially malicious external URL detected (high risk: 0.80). The script instructions suggest running curl -fsSL https://bibigpt.co/install.sh | bash to install the CLI binary, which is a remote code execution risk via external script installation.