Web Distiller for Openclaw

Web Distiller converts raw webpage URLs into clean, LLM-optimized Markdown or text to streamline AI agent data ingestion.

edwardyen724-g
v0.1.0
Mar 23, 2026
0
730
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install web-distiller

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install web-distiller 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 Web Distiller?

Web Distiller is a specialized utility designed to bypass the noise of raw HTML and deliver high-quality, cleaned content directly to your AI agents. By utilizing the Distiller API, this skill ensures that agents receive structured Markdown instead of cluttered webpage fetches, significantly reducing prompt token usage and improving context accuracy. As part of the Openclaw Skills library, it prioritizes reliability and developer experience by providing fallback mechanisms and clear setup paths.

The skill acts as a bridge between the live web and your AI workflows, offering flexible endpoints tailored to different needs. Whether you require a simple text summary or a complex distillation of a JavaScript-heavy site, Web Distiller handles the heavy lifting of content extraction and formatting.

Web Distiller Use Cases

  • Extracting clean documentation from technical websites for RAG (Retrieval-Augmented Generation) systems.
  • Automating bulk data collection from multiple URLs for competitive research or market analysis.
  • Reducing token costs by stripping HTML boilerplate and ads before passing content to an LLM.
  • Scouring JavaScript-heavy single-page applications using headless browser rendering options.

How Web Distiller Works

  1. The operator provides a target URL and specifies the desired output format (Markdown, Text, or JSON).
  2. The skill checks for a valid DISTILLER_API_KEY; if missing, it guides the user through the authentication process at webdistiller.dev.
  3. A request is sent to the preferred endpoint, defaulting to /markdown for broad compatibility or /distill for advanced cleaning on paid plans.
  4. The Distiller API processes the page, removing noise and formatting the core content.
  5. The cleaned data is returned to the agent workflow, ready for immediate processing or storage.

Web Distiller Setup

Install the package via pip:

pip install web-distiller

Configure your environment variables in your .env file or shell:

export DISTILLER_API_BASE=https://webdistiller.dev
export DISTILLER_API_KEY=your-api-key

If you do not have a key, sign in at [https://webdistiller.dev/signin](https://webdistiller.dev/signin) and retrieve it from your dashboard.

Web Distiller Data Schema & Taxonomy

Web Distiller supports multiple output schemas to fit different technical requirements:

Format Description Best Use Case
Markdown Structured content with headers and links. LLM Prompts & Documentation
Text Raw plain text with minimal formatting. Smallest prompt payload
JSON Full object including metadata and billing info. Tool integration & Metadata tracking

For batch operations, the skill returns a job_id which can be used to poll the status and retrieve a collection of processed results.

Web Distiller Advanced Features

  • Multi-mode Batch Processing: Handle hundreds of URLs simultaneously using the /batch endpoint.
  • Browser Rendering: Use the --use-browser flag to execute JavaScript for dynamic web applications.
  • Dynamic Fallbacks: Automatically reverts from /distill to /markdown if subscription limits are met to ensure workflow continuity.
  • Flexible Endpoints: Native support for both free-tier and premium extraction modes tailored for the Openclaw Skills framework.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*