A robust tool for converting websites into structured JSON or clean Markdown using the Tabstack API and Babashka.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install tabstack-extractor
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 tabstack-extractor 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 Tabstack Extractor?
This skill empowers AI agents to perform high-fidelity web scraping by interfacing with the Tabstack API. It bridges the gap between raw HTML and structured data, allowing for consistent extraction of job listings, product details, or news articles based on predefined JSON schemas.
By leveraging Openclaw Skills, developers can automate the process of turning the web into a queryable database. The skill is built with Babashka, ensuring lightning-fast startup times and a robust environment for complex data extraction workflows without the overhead of heavy dependencies. It is designed to be highly reliable, featuring built-in retry logic and caching to streamline data gathering tasks.
Tabstack Extractor Use Cases
- Scraping job listings from LinkedIn or company career pages with specific fields like salary and requirements.
- Converting messy blog posts or news articles into clean Markdown for LLM analysis or summarization.
- Extracting e-commerce product data, including pricing, descriptions, and images, into structured JSON.
- Gathering contact information or metadata from corporate websites for research and lead generation.
- Performing batch extraction across multiple URLs to build large-scale datasets for analysis.
How Tabstack Extractor Works
- The user provides a target URL and optionally a JSON schema defining the desired data structure.
- The skill authenticates with the Tabstack API using a secure API key stored in environment variables or a configuration file.
- A Babashka script sends a request to Tabstack, which processes the webpage and extracts data according to the schema or converts it to Markdown.
- The skill applies built-in retry logic and caching mechanisms to ensure reliability and minimize API costs.
- The structured JSON or Markdown result is returned to the AI agent or saved to a local file for further processing within the Openclaw Skills ecosystem.
Tabstack Extractor Setup
Install Babashka to run the extraction scripts effectively:
curl -s https://raw.githubusercontent.com/babashka/babashka/master/install | bash
Configure your Tabstack API key as an environment variable to allow Openclaw Skills to authenticate:
export TABSTACK_API_KEY="your_api_key_here"
Test your connection to verify the setup:
bb scripts/tabstack.clj test
Tabstack Extractor Data Schema & Taxonomy
The skill utilizes JSON schemas to define the structure of the extracted data, ensuring that every scrape results in a consistent format. This is particularly useful for Openclaw Skills that require predictable data inputs.
| Feature | Description |
|---|---|
| JSON Schema | Defines keys like title, price, description, and requirements. |
| Markdown | Provides a stripped-down, readable version of the page content. |
| Cache | Local storage for 24-hour retention of previous API results. |
| Batch File | A simple text file containing a list of URLs for bulk processing. |
Tabstack Extractor Advanced Features
- Intelligent retry logic with 3 retries and 1s delay to handle transient network errors.
- Local 24-hour caching to prevent redundant API calls and reduce latency for repeated requests.
- Parallel batch processing for high-volume URL extraction from text files.
- Pre-built schema templates for jobs, news, and products located in the references directory for quick starts.
- Seamless integration with other Openclaw Skills like web search or browser automation tools for complex multi-step workflows.
SKILL.md
Loading
Related Openclaw Skills
SearXNG Web Search
A privacy-preserving web search skill for AI agents that leverages self-hosted SearXNG instances to retrieve real-time data and documentation.
Tsundere Persona
An AI skill that transforms standard model responses into classic tsundere anime character dialogue.
Anime Meme Collector
An automated system for harvesting and organizing Chinese ACG memes and trending Bilibili content to improve AI cultural intelligence.
DHT11 Temperature & Humidity Sensor
A specialized tool for reading real-time climate data from DHT11 sensors via Raspberry Pi GPIO pins.
pi-speaker
A local audio playback workflow that enables AI agents to play speech and sound files directly through a Raspberry Pi or gateway host speaker.
Apollo Issue Review
A professional issue triage workflow that automates reproduction, evidence-checking, and drafting maintainer-grade GitHub responses for the Apollo ecosystem.
Featured*
