Remote Visual Browser Deployment
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md and install @user_15292d5a/yjkj-remote-browser-deploy.
About this skill
Problem to Solve
When an AI agent opens pages, clicks, screenshots, or logs in, a headless browser is invisible to the user. Human handoff for sliders, QR codes, SMS verification, and shared debugging becomes difficult. remote-browser-deploy addresses this visible, shared-control need: the agent and user view the same browser session.
How It Works and Key Steps
The skill branches by OS. On Linux, it uses Xvfb as a virtual display, runs Chromium, and exposes the screen through x11vnc, noVNC, and websockify on 6080/tcp. The agent controls the browser through CDP on 9223/tcp. On Windows, it reuses local Edge or Chrome and enables CDP without extra VNC components. After deployment, it checks port listeners, vnc.html availability, and CDP json page data.
Important constraints: use the CDP browser deployed by this skill, do not add --headless, select the correct address based on CLOUD, NAS, or LOCAL, and send core completion info before the final screenshot.
Boundaries
It fits login consoles, risk verification, content publishing, and web debugging. It is not suitable for pure API tasks, high-security production environments where remote desktop access must not be exposed, or Linux servers that cannot open 6080/tcp or lack root/sudo permissions.
Use Cases
- Debug a Douyin creator-center publish page where the agent fills fields while a user handles slider verification.
- Inspect a Xiaohongshu login-page issue by opening the page in an agent-controlled browser visible through noVNC.
- Deploy a visible Chromium instance on a Linux cloud server so the agent can control it via CDP and capture screenshots.
- Debug a WeChat official-account QR-code flow when the user and agent must view the same browser state.
Best For
- Automation engineers who need an agent to log into creator consoles while humans handle verification
- Operations engineers who run visible Chromium on headless Linux servers
- Frontend debugging engineers troubleshooting login flows, captchas, and inconsistent page state
- Automation integrators who use AI agents to control browsers but require a shared visible session
Related Skills
Automatically indexes Gradle-cached AAR/JAR dependency classes and returns library coordinates, versions, and public APIs by fully qualified name, using only the Python standard library.
Codifies AMT and YourMT3 training conventions, script patterns, hyperparameters, precision, checkpoints, and NaN safeguards.
Retrieve relevant chunks from a customer-managed PKM dataset by dataset_id and return concise, source-annotated answers.
Convert PRDs, user stories, or functional specs into prioritized test-point checklists covering functional, business-rule, boundary, exception, and non-functional dimensions.