An AI-powered web scraping and browser automation agent skill that integrates AgentQL with OOMOL CLI to query, extract, and manipulate web page data without local credential management.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install oo-agentql
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 oo-agentql 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 AgentQL OOMOL Connector?
The AgentQL OOMOL Connector is a high-performance skill designed to integrate AgentQL's next-generation semantic web scraping with OOMOL-connected accounts. By leveraging this tool within Openclaw Skills, developers can interactively orchestrate headless browsers, create remote Tetra browser sessions, and retrieve structured data from any webpage using AI-driven selectors. The execution is handled completely server-side via the OOMOL command-line interface, meaning developers never have to worry about managing raw API keys or exposing credential tokens.
Integrating this capability allows AI agents to effortlessly read, create, and update web data structures. This is particularly valuable for web automation pipelines, dynamic data harvesting, and maintaining automated QA flows. By wrapping AgentQL with Openclaw Skills, your AI workflow gains robust, secure web interaction capabilities that can adapt dynamically as websites change.
AgentQL OOMOL Connector Use Cases
- Extracting structured semantic data from highly dynamic, JavaScript-heavy websites.
- Provisioning remote headless browser sessions via AgentQL Tetra for remote debugging or visual testing.
- Monitoring account API usage and tracking browser session telemetry for billing and analytics.
- Automating multi-step form submissions or interactive workflows behind logins.
How AgentQL OOMOL Connector Works
- Retrieve Action Schema: The AI agent queries the live schema of the desired AgentQL action using the OOMOL CLI to ensure structural compliance.
- Construct JSON Payload: The agent prepares a JSON-formatted payload based on the retrieved schema, tailoring it to the target website and target parameters.
- Execute via OOMOL CLI: The command is dispatched using the CLI, leveraging server-side credential injection to securely authenticate with AgentQL.
- Parse Structured Response: The agent receives structured data (and an execution ID) directly from the AgentQL API, making it immediately available for downstream processing.
AgentQL OOMOL Connector Setup
These are one-time setup steps. Run a step only when a command fails for the matching reason.
1. Install OOMOL CLI
If the CLI is not yet installed, download and install it based on your operating system:
For macOS / Linux:
curl -fsSL https://cli.oomol.com/install.sh | bash
For Windows PowerShell:
irm https://cli.oomol.com/install.ps1 | iex
2. Sign In to OOMOL
Sign into your account once to enable command execution:
oo auth login
3. Connect AgentQL
Navigate to the OOMOL Console to connect your AgentQL account using an API key:
https://console.oomol.com/app-connections?provider=agentql
AgentQL OOMOL Connector Data Schema & Taxonomy
The skill operates on structured inputs and outputs fetched dynamically from the OOMOL schema registry. Below is the structure of the primary actions and data schemas:
Core Actions
| Action Name | Description | Tag |
|---|---|---|
create_browser_session |
Create an AgentQL Tetra remote browser session | [write] |
query_data |
Query a webpage and return extracted structured data | Read (Safe) |
get_usage |
Fetch AgentQL account and API key usage counters | Read (Safe) |
list_session_usage |
List AgentQL Tetra session telemetry entries | Read (Safe) |
CLI Command Reference
To inspect any live action schema before dispatching payloads, use:
oo connector schema "agentql" --action "<action_name>"
Execution payloads return responses mapped inside a root key:
{
"data": {
"result": "<extracted_structured_data>"
},
"meta": {
"executionId": "<execution_id_string>"
}
}
AgentQL OOMOL Connector Advanced Features
- Dynamic Action Inspection: Fetch authoritative input/output schemas programmatically before execution to prevent structural payloads from failing.
- Tetra Remote Browser Sessions: Easily initialize complete browser remote control flows with automated session telemetry.
- Granular Safety Controls: Tagged operations guarantee security, explicitly flagging write and destructive actions so agents can prompt for confirmation.
- Server-side Token Isolation: Keep API keys fully protected without exposing credential states in the client code.
- Flexible Orchestration with Openclaw Skills: Seamlessly combine with other open-source tools and platforms using Openclaw Skills to design custom data mining pipelines.
SKILL.md
Loading
Related Openclaw Skills
Ably Integration Skill
A powerful OOMOL-powered skill to manage Ably real-time messaging, channels, presence history, and push notifications directly from your AI coding environment.
Amara Subtitling and Video Management
Automate Amara subtitle tracks, video URLs, and translation pipelines securely using the oo CLI and agent-friendly schemas.
oo-apiverve
A versatile skill enabling AI agents to interact with APIVerve's suite of utility APIs, including calculations, lookups, and environmental data fetching.
BuiltWith (oo-builtwith)
A powerful integration skill for the oo CLI that enables AI agents to query BuiltWith domain profiles, tech stacks, and social media mappings instantly.
Best Buy
An AI agent skill to programmatically search products, read store details, and retrieve customer reviews from Best Buy using the secure oo CLI.
Nobody's Girl Memoir Analysis Skill
An AI-powered researcher skill that provides deep insight and analysis on Virginia Roberts Giuffre's posthumous memoir, Nobody's Girl, detailing her survival and fight for justice.
Featured*
