agentql-automation

Automate Agentql tasks via Rube MCP (Composio). Always search tools first for current schemas.

Install

Hot:2

Download and extract to your skills directory

Copy command and send to OpenClaw for auto-install:

Download and install this skill https://openskills.cc/api/download?slug=composiohq-composio-skills-agentql-automation&locale=en&source=copy

AgentQL Automation Skills — AgentQL Task Automation via Rube MCP

Overview


AgentQL Automation is a Claude skill that enables AgentQL task automation through Composio’s Rube MCP server, allowing you to perform web queries, data collection, and batch operations without API keys.

Use Cases

1. Automated Web Data Extraction


When you need to extract structured data in bulk from multiple web pages, this skill lets you quickly obtain the target information using AgentQL’s natural-language queries—without writing complex selectors or scraper code.

2. Automated Web Interaction Tasks


For scenarios that require simulating user actions on web pages (such as filling out forms, clicking buttons, or scrolling), this skill provides a complete toolchain to support automation for testing, data entry, and more.

3. Batch Operations Across Multiple Sessions


When you need to manage multiple AgentQL tasks or workflows at the same time, this skill’s session reuse and multi-tool execution capabilities can significantly improve efficiency. It supports paginated data processing and complex workflow orchestration.

Core Features

1. Smart Tool Discovery


By using RUBE_SEARCH_TOOLS, the system automatically fetches the currently available AgentQL tool list and the latest input schema—avoiding compatibility issues caused by hard-coding tool names. The system returns a recommended execution plan and known pitfalls to help you use the tools correctly.

2. Connection Management and Verification


Use RUBE_MANAGE_CONNECTIONS to check the AgentQL connection status. If the connection is not active, the system provides an authentication link to complete the setup. Connection status is validated before every tool execution to ensure reliable task runs.

3. Coordinated Multi-Tool Execution


With RUBE_MULTI_EXECUTE_TOOL, you can execute multiple AgentQL tools in a single call. Combined with session ID reuse, it supports complex workflows. It supports pagination handling and result aggregation, making it suitable for large-scale batch data processing.

FAQ

Does the AgentQL Automation skill require an API key?


No. The Rube MCP server is an API-key-free MCP service—just add https://rube.app/mcp as the server endpoint in your client configuration. However, AgentQL itself requires a one-time OAuth authorization of the account via Composio.

How do I get the correct schema for AgentQL tools?


Be sure to call RUBE_SEARCH_TOOLS before performing any operation, passing in your specific use case. Tool schemas are updated frequently, and hard-coding parameters can easily cause calls to fail. Search results include the complete input parameter format and a recommended execution plan.

What should I do if the connection status shows “inactive”?


Use RUBE_MANAGE_CONNECTIONS to check the connection status. If it returns a non-ACTIVE state, the system will provide an authentication link. After clicking the link to authorize the Composio AgentQL tool package, check the status again to confirm it becomes ACTIVE so you can use it normally.

Are there any limitations for this skill?


This skill depends on the availability of the Rube MCP service and the functional scope of the Composio AgentQL tool package. For very large-scale batch tasks, it’s recommended to use the run_composio_tool() method of RUBE_REMOTE_WORKBENCH for better performance. Also, ensure correct session ID reuse to avoid creating too many separate sessions.