xiaohongshu-ops

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

• Third-party content exposure detected (high risk: 0.90). The skill (SKILL.md and referenced files like references/xhs-viral-copy-flow.md and references/xhs-home-feed-analysis.md) explicitly instructs the agent to use the built-in browser to fetch and scrape public Xiaohongshu posts, comments and arbitrary note URLs (user-generated content) and to analyze those pages to drive decisions and publishing, which exposes the agent to untrusted third-party content that could contain indirect prompt injections.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill's recommended one‑click install script calls curl -fsSL https://raw.githubusercontent.com/nvm-sh/nvm/v0.40.0/install.sh | bash at runtime, which fetches and executes remote code (nvm) as part of the required OpenClaw installation for this skill.