小红书全能运营助手，AI生图 + 草稿发布

• Risk level: Low
• Analyzed: Apr 3, 2026, 9:25 AM
• Findings: 0
• Version: 1.0.0

Overview

The skill's code, docs, and runtime instructions are coherent with its stated purpose (automating AI image generation and automating Xiaohongshu draft publishing), but it requires controlling a local browser via CDP which carries privacy risks the user should understand.

This skill appears to do what it claims: generate images (via MiniMax API or by automating a web service) and automate draft uploads to Xiaohongshu by controlling a locally running OpenClaw/Playwright browser over CDP (default port 18800). Before installing/using: 1) Review scripts (upload_auto.js, generate_image_smart.js) yourself — they control your local browser and can interact with any open page and its logged-in sessions. 2) Run in a dedicated/browser profile or isolated environment so the automation cannot access other browser accounts or sensitive tabs. 3) Check or create scripts/config.json yourself — the MiniMax API key is stored there; do not put other secrets in that file. 4) If you are uncomfortable with automated browser control, do not run the scripts or only run them after auditing the code. 5) When in doubt, run the commands manually step‑by‑step (inspect screenshots in the temp folder) rather than allowing unattended/autonomous runs. I have medium confidence because parts of the scripts were truncated in the provided listing — full review of all code paths (especially the truncated sections that download/extract URLs) would raise confidence to high.