himalaya

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

• Third-party content exposure detected (medium risk: 0.30). The Himalaya CLI skill configures an AI agent to read, list, and process emails via IMAP/SMTP where untrusted external parties can send arbitrary free text bodies.

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

• Potentially malicious external URL detected (high risk: 0.80). The installation instructions use a curl command to fetch and execute a remote shell script from GitHub (https://raw.githubusercontent.com/pimalaya/himalaya/master/install.sh), which represents a remote code execution dependency.