A browser-based automation skill that enables OpenClaw agents to perform deep interactions with the Gemini web interface, including authentication and asset management.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install openclaw-gemini-web
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install openclaw-gemini-web 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 OpenClaw Gemini Web Skill?
The openclaw-gemini-web skill is a specialized component within the Openclaw Skills ecosystem designed to control the Gemini web interface through automated browser tools. Unlike API-based integrations, this skill leverages the full visual and functional capabilities of the Gemini web platform, allowing agents to manage threads, upload files for analysis, and trigger image generation just as a human user would. This approach is ideal for workflows that require features exclusive to the web version of Gemini.
By utilizing this skill, developers can maintain persistent login states within managed browser profiles, reducing the friction of repetitive authentication. It bridges the gap between local file systems and cloud-based AI, ensuring that outputs like generated images or summarized text are captured and organized locally for immediate use in larger automation pipelines.
OpenClaw Gemini Web Skill Use Cases
- Automating the login process to Gemini using stored credentials and 2FA secrets.
- Uploading local documents, such as PDFs or datasets, for Gemini to analyze and summarize.
- Generating high-quality images via the Gemini interface and automatically downloading them to a local project folder.
- Branching or continuing long-running chat threads to explore different reasoning paths or creative directions.
- Conducting browser-assisted research where the agent needs to synthesize information directly from the Gemini web UI.
How OpenClaw Gemini Web Skill Works
- The agent launches a managed browser instance specifically configured for Openclaw Skills to maintain isolation from personal profiles.
- It navigates to the Gemini web portal and assesses the current authentication state.
- If unauthenticated, the skill attempts an automated login using provided environment variables, including real-time TOTP generation for two-factor authentication.
- Once the dashboard is ready, the skill executes the primary task, such as sending prompts, uploading attachments, or switching threads.
- The skill monitors the interface for completion, ensuring that long-form responses or image generations are fully rendered before proceeding.
- Finally, it captures any generated assets, downloads them to a structured local directory, and standardizes file naming for consistency.
OpenClaw Gemini Web Skill Setup
To get started with this skill in your Openclaw Skills environment, ensure you have python3 installed and set the required environment variables for authentication.
# Configure your Gemini credentials
export GEMINI_WEB_EMAIL="[email protected]"
export GEMINI_WEB_PASSWORD="your-secure-password"
# Optional: Configure TOTP for automated 2FA
export GEMINI_WEB_TOTP_SECRET="YOUR_BASE32_SECRET"
You can test the TOTP generation script included in the skill directory:
python3 scripts/totp.py --env GEMINI_WEB_TOTP_SECRET
OpenClaw Gemini Web Skill Data Schema & Taxonomy
The skill organizes its output into a standardized directory structure to ensure that all assets retrieved via Openclaw Skills are easy to locate and process.
| Path | Description |
|---|---|
./output/gemini/YYYY-MM-DD/ |
The root directory for all assets generated on a specific date. |
session-notes.md |
A summary file containing context about the chat session and metadata. |
[name]-[index].[ext] |
Downloaded assets (images, PDFs) renamed to a clean, lowercase hyphenated format. |
scripts/totp.py |
Utility script for generating 2FA codes during the login phase. |
OpenClaw Gemini Web Skill Advanced Features
- Multi-source TOTP support: Generate authentication codes from Base32 secrets, otpauth URIs, or JSON-stored credentials.
- Human-in-the-loop fallback: Automatically pauses and alerts the user if CAPTCHA or advanced security challenges are detected.
- Intelligent thread management: Supports branching existing conversations (forking) to preserve context while starting new tasks.
- Automated asset pipelines: Downloads and renames generated images and files into stable local paths for further processing.
- Contextual file uploads: Handles the sequential process of uploading reference materials before initiating complex prompts.
SKILL.md
Loading
Related Openclaw Skills
System Cleanup
A specialized utility to audit and safely remove system junk, caches, and logs on Ubuntu and WSL environments.
Research Queue
An autonomous background research system that captures and investigates technical, product, and algorithmic questions using structured queues.
Gate Exchange Internal Transfer
A professional skill for executing and automating internal fund transfers between different account types within the Gate Exchange ecosystem.
Gate Unified Account Assistant
A specialized AI skill for managing Gate.io unified account equity, margin borrowing, and leverage configurations with high-precision data handling.
Buffett Investment Advisor
An AI-driven investment consultant distilled from seven decades of Warren Buffett's shareholder letters and core value investing principles.
AI Viral Trend Hijacker
A real-time trend detection and automated video production engine designed to hijack viral social media momentum before it peaks.
Featured*
