clawhub-hardened

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to fetch and install skills from a public registry (default registry: https://clawhub.com) via commands like clawhub install / clawhub update, meaning the agent will ingest and potentially execute untrusted third‑party skill code from the open web (clawhub.com), which can materially influence its actions.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly uses the ClawHub registry at https://clawhub.com as the default runtime source for installing/updating skills via the clawhub CLI, which will fetch remote skill code/content that can control agent prompts or execute code.