A browser automation skill that enables AI agents to query Grok AI for real-time information and sophisticated multi-turn research.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install grok-query
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 grok-query 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 Grok External Knowledge Query?
The grok-query skill is a specialized component of Openclaw Skills designed to give AI agents access to real-time, external knowledge via the grok.com interface. Unlike standard search tools, this skill uses browser automation to interact with Grok's advanced reasoning capabilities, allowing for deep synthesis and context-aware answers. It is ideal for scenarios where the agent needs information that post-dates its training data or requires the specific logical nuance provided by the Grok model.
By automating the navigation, interaction, and extraction phases, Openclaw Skills ensures that agents can retrieve high-quality data while handling complex UI elements like overlays and dynamic input boxes. This skill transforms a standard AI agent into a real-time researcher capable of browsing the web through the lens of another powerful AI model.
Grok External Knowledge Query Use Cases
- Retrieving the latest breaking news, product launches, or real-time event updates that are not available in static datasets.
- Conducting deep-dive research that requires multi-turn conversational context to refine answers.
- Verifying uncertain technical facts or current software versioning through the Grok interface.
- Performing reasoning tasks that exceed the capabilities of standard web search results.
How Grok External Knowledge Query Works
- The agent utilizes Openclaw Skills to either locate an existing grok.com tab or open a new one.
- A page snapshot is captured to identify and dismiss blocking UI elements like SuperGrok upgrade banners or cookie consents.
- The agent identifies the specific input element, which is typically a contenteditable div at the bottom of the page.
- The query is typed into the input field, followed by a strategic click on the Submit button, bypassing the newline-only behavior of the Enter key.
- The agent enters a polling loop, taking periodic snapshots to detect completion indicators such as the Regenerate button.
- Once the response is finished, the agent extracts the text or image URLs from the DOM.
Grok External Knowledge Query Setup
To use this skill, ensure the browser environment is configured and you are manually logged into your Grok account.
# Verify browser readiness
openclaw browser status
# Open the Grok interface if no tab exists
openclaw browser open https://grok.com
Grok External Knowledge Query Data Schema & Taxonomy
The skill manages interaction data through the following taxonomy:
| Element | Description |
|---|---|
| Tab ID | The persistent identifier for the grok.com session. |
| Snapshot Refs | Dynamic references for input fields and buttons used within a single turn. |
| State Indicators | UI markers such as Stop response or Regenerate to track model progress. |
| Media URLs | Source links for any images generated by Grok during the session. |
Grok External Knowledge Query Advanced Features
- Smart overlay handling that uses the Escape key or targeted clicks to clear UI obstructions.
- Multi-turn conversation persistence, allowing the agent to reference previous messages in the same tab.
- Image extraction capabilities to handle Grok's multimodal response outputs.
- Timeout-resilient polling logic that avoids tool-call failures during long generation processes.
- Localized UI support to handle various language settings on the grok.com platform.
SKILL.md
Loading
Related Openclaw Skills
Employee Skills Importer
Automate the transformation of complex employee skill CSV files into clean, idempotent SQL scripts for Supabase.
Autotask MCP for Kaseya PSA
A local Docker-based Model Context Protocol (MCP) server to bridge AI agents with Datto/Kaseya Autotask PSA for ticket and project management.
Trench
Trench is a high-performance execution skill for AI agents to trade Solana meme coins with MEV protection and automated rug detection.
Auto Memory Curation
A silent intelligence layer that analyzes every interaction to build a structured, long-term memory of project facts and user preferences.
Kmoe Manga Downloader
A powerful CLI skill for downloading manga from Kmoe in EPUB or MOBI formats with advanced credential management and automation.
Instagram Marketing Generator
An automated tool for converting e-commerce product URLs into optimized Instagram marketing packages including captions, hashtags, and visual briefs.
Featured*
