Distil enables AI agents to search the web and fetch web pages as clean, LLM-friendly Markdown to drastically reduce token consumption.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install distil
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 distil 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 Distil?
The Distil skill provides agents with a streamlined, discoverable way to access the Distil proxy service. This tool is a cornerstone of the Openclaw Skills ecosystem for developers who need to feed high-quality web data into Large Language Models without the noise of raw HTML. By converting web pages into optimized Markdown, Distil typically saves between 60% and 80% of tokens, making it both cost-effective and more readable for AI agents.
In addition to simple page fetching, this skill enables full-scale web searches, JavaScript rendering for modern web apps, and visual screenshotting. It removes the friction of manual URL construction and header management, allowing the agent to focus on information synthesis rather than data cleaning.
Distil Use Cases
- Reducing token costs by converting cluttered HTML documentation into clean Markdown.
- Performing real-time web research to ground AI responses in current facts.
- Scraping single-page applications (SPAs) that require JavaScript rendering to display content.
- Capturing visual screenshots of web pages for multimodal analysis.
- Bypassing local caches to ensure the agent is reading the most recent version of a news site or repository.
How Distil Works
- The AI agent identifies a target URL or a search query needed to complete a task.
- The skill constructs a request to the Distil proxy using the configured environment variables provided by Openclaw Skills.
- Distil fetches the remote content, optionally rendering JavaScript if the specific endpoint is called.
- The service strips away navigation bars, ads, and boilerplate, transforming the core content into Markdown.
- The cleaned content is returned to the agent's context, including metadata about how many tokens were saved during the conversion.
Distil Setup
To integrate this component of Openclaw Skills into your workflow, follow these steps:
- Register for a free API key at https://distil.net.
- Set the required environment variable in your terminal or agent configuration:
export DISTIL_API_KEY="your_api_key_here"
- (Optional) If you are using a self-hosted version, set the proxy URL:
export DISTIL_PROXY_URL="https://proxy.distil.net"
Distil Data Schema & Taxonomy
The Distil skill interacts with the proxy to return data in various formats depending on the endpoint used. The output is typically sent to stdout for the agent to consume.
| Endpoint | Output Format | Purpose |
|---|---|---|
/https://url |
Markdown | Standard cleaned web content |
/search?q=... |
Markdown | Search engine results converted to Markdown |
/screenshot/ |
PNG Image | Visual capture of the web page |
/render/ |
Markdown | Content extracted after full JS execution |
/raw/ |
HTML/Text | Original source code without processing |
Distil Advanced Features
- Real-time token efficiency reporting included in Markdown responses to monitor optimization performance.
- Support for custom proxy URLs, allowing for seamless integration with self-hosted instances within Openclaw Skills.
- No-cache mode using the
/nocacheprefix to force a fresh fetch from the source server. - Automatic fallback to raw HTML if token limits for Markdown conversion are exceeded, ensuring no loss of information.
- Native support for screenshotting, enabling agents to process visual layouts of websites.
SKILL.md
Loading
Related Openclaw Skills
DeShell
A high-performance proxy skill that converts web pages and search results into LLM-friendly Markdown to optimize token consumption.
Governance Wrapper
A mandatory security wrapper for strict-mode execution, evidence logging, and policy enforcement in autonomous operations.
OctoFlow
A zero-dependency, GPU-native programming language that converts natural language descriptions into high-performance Vulkan compute programs.
Wisdom & Accountability Coach
A sophisticated AI mentor that uses longitudinal memory and philosophical frameworks to guide personal development and maintain compassionate accountability.
Late API Social Media Integration
A comprehensive API skill for scheduling and managing social media content across 13 major platforms from a single interface.
Late CLI Social Media Scheduler
Late CLI is a powerful command-line interface for scheduling and publishing social media content across 13 major platforms from any terminal or AI agent.
METADATA
Featured*
