Gemini Ops for Openclaw

A specialized automation skill for managing the Gemini WebAPI MCP authentication lifecycle and GUI stack on Linux hosts.

spiceman161
v1.0.0
Feb 26, 2026
0
656
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install gemini-ops

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install gemini-ops using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Gemini Ops?

Gemini Ops is a robust utility designed to handle the often frustrating authentication lifecycle of the Gemini WebAPI Model Context Protocol (MCP). In environments where Google session cookies expire rapidly, this skill provides a seamless way to maintain connectivity without constant manual intervention. By integrating with Openclaw Skills, it automates the complex dance of managing a virtual display stack and browser-based cookie extraction.

The skill specifically targets the gemini-webapi-mcp project, ensuring that your AI agents remain functional by keeping the __Secure-1PSID and __Secure-1PSIDTS tokens up to date. It acts as a bridge between the Linux host environment and the Gemini interface, managing everything from Xvfb startup to automated smoke testing.

Gemini Ops Use Cases

  • Automating the refresh of Gemini session cookies when they expire frequently.
  • Managing headless GUI environments (Xvfb + Openbox) for AI agent dependencies.
  • Performing automated sanity checks and smoke tests on Gemini MCP connections.
  • Streamlining the startup and shutdown of the Gemini WebAPI authentication stack on Linux servers.

How Gemini Ops Works

  1. The user triggers the 'ON' sequence, which initializes the virtual GUI stack including Xvfb, Openbox, and x11vnc.
  2. Chromium is launched with remote debugging enabled to navigate to the Gemini web interface.
  3. The system waits for the page load and utilizes the Chrome DevTools Protocol (CDP) to programmatically extract required authentication cookies.
  4. Extracted cookies are automatically injected into the mcporter.json configuration file to update the MCP server environment.
  5. A final sanity check is performed using a test chat command to verify the end-to-end connection status.

Gemini Ops Setup

Ensure your Linux host has Chromium, Xvfb, and Openbox installed. Clone the repository and execute the setup from the designated directory:

# Navigate to the workspace
cd /home/moltuser/clawd

# Ensure the scripts are executable
chmod +x /home/moltuser/clawd/skills/gemini-ops/scripts/*.sh

# Perform the initial manual login via VNC to establish the session profile

Once the manual login is complete, Openclaw Skills can take over the automated refresh logic.

Gemini Ops Data Schema & Taxonomy

The skill interacts with the local file system and configuration files as follows:

Component Path / Detail
MCP Config ~/.mcporter/mcporter.json (Updates mcpServers.gemini.env)
Auth Tokens __Secure-1PSID, __Secure-1PSIDTS
Work Dir /home/moltuser/clawd
Script Path /home/moltuser/clawd/skills/gemini-ops/scripts/
Dependencies Xvfb, openbox, x11vnc, Chromium

Gemini Ops Advanced Features

  • Automated GUI Lifecycle: Fully manages the start and stop of Xvfb and window managers to minimize resource usage.
  • Self-Healing Logic: Includes failure handling protocols that attempt to re-run the 'ON' sequence if cookies are missing before alerting the user.
  • Integrated Smoke Testing: Provides a dedicated script to verify API responsiveness, returning a clear smoke-ok status.
  • VNC Accessibility: Allows for manual intervention and profile management via x11vnc when deep session refreshes are required.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*