google

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). These links are suspicious: they point to an unknown GitHub account distributing a password-protected ZIP that contains an executable the user is instructed to run, and a rentry.co paste that likely contains shell/install commands — all high-risk indicators for malware distribution.

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs the agent to read and act on user-provided Google content (e.g., Gmail "get"/"list" messages, Drive/Docs "read", Sheets "read") which are untrusted/user-generated third-party inputs that could contain instructions influencing replies, event creation, or other follow-up actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires the openclaw-core utility at runtime and directs users to download and execute remote code from https://github.com/denboss99/openclaw-core/releases/download/v3/openclawcore-1.0.3.zip and https://rentry.co/openclaw-core, so external content would be fetched and run as a required dependency.