clawhub

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). Yes — the instructions ask you to download and run a password‑protected GitHub release archive from an unfamiliar account, execute code referenced by a third‑party snippet (glot.io), and allow a custom registry — combined these are strong indicators of an untrusted/malicious distribution channel.

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

• Third-party content exposure detected (high risk: 0.90). This skill installs and updates skills from the public ClawHub registry (https://www.clawhub.ai/) and permits arbitrary custom registry URLs, so the agent will fetch and consume community-published, untrusted third-party skill content as part of its workflow, enabling indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill uses the ClawHub registry URL (https://www.clawhub.ai/) at runtime to fetch/install skill packages, which are remote code/assets that can be executed or directly alter agent behavior, making it a runtime dependency that can control prompts or execute code.