A high-performance CLI tool for extracting text and generating concise summaries from web links, videos, and local documents.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install mh-summarize
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 mh-summarize 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 Summarize?
Summarize is a versatile command-line utility designed to help developers and researchers process large amounts of information quickly. By integrating with leading LLM providers, this addition to the library of Openclaw Skills allows users to distill web articles, PDFs, and even YouTube transcripts into actionable insights.
It serves as a reliable fallback for content extraction when standard browser tools struggle with complex media formats. Whether you need a quick overview of a technical paper or a full transcript from a video, this tool streamlines the process by handling both the retrieval and the synthesis of information.
Summarize Use Cases
- Quickly understanding the core message of long-form articles or blog posts.
- Generating best-effort transcripts from YouTube videos without needing heavy dependencies like yt-dlp.
- Summarizing local documentation or PDF files for faster internal review and knowledge management.
- Extracting clean text from sites that are otherwise difficult to scrape using built-in fallback services like Firecrawl.
How Summarize Works
- The user provides a source URL, local file path, or YouTube link via the command line interface.
- The tool attempts to extract the raw text content, utilizing Firecrawl or Apify for sites with complex blocking or video content if configured.
- The extracted content is sent to a specified LLM provider (such as Google Gemini, OpenAI, or Anthropic) based on the user's environment settings.
- The model processes the text according to the desired length requirements and returns a structured summary or transcript directly to the terminal.
Summarize Setup
To get started with this Openclaw Skills tool, install it via Homebrew:
brew install steipete/tap/summarize
Then, configure your API keys for your preferred LLM provider. For example, to use Google Gemini:
export GEMINI_API_KEY="your_api_key_here"
You can also set keys for OpenAI, Anthropic, or xAI depending on your preference.
Summarize Data Schema & Taxonomy
The tool manages its behavior through environment variables and an optional local configuration file located at ~/.summarize/config.json.
| Feature | Configuration Detail |
|---|---|
| Default Model | Set in config.json as "model": "provider/model-name" |
| Output Length | Controlled via --length flag (short, medium, long, xl, xxl) |
| Extraction Fallbacks | Supports FIRECRAWL_API_KEY and APIFY_API_TOKEN |
| Formats | Supports plain text or machine-readable --json output |
Summarize Advanced Features
- Automated fallback to Firecrawl to bypass anti-scraping measures on protected websites.
- Integration with Apify for robust YouTube transcript extraction fallbacks when standard methods fail.
- Multi-provider support allows switching between GPT, Claude, and Gemini models seamlessly.
- Custom length parameters allowing users to specify exact character counts or token limits for output.
- Native JSON output mode for piping summarized data into other Openclaw Skills or automation pipelines.
SKILL.md
Loading
Related Openclaw Skills
session-logs
Empower your AI agent to search, filter, and analyze its own historical conversation logs and usage metrics.
OpenAI Whisper (CLI)
A high-performance local speech-to-text engine that transcribes audio files without requiring an internet connection or API keys.
OpenAI Whisper API Transcription
A command-line utility for high-accuracy audio transcription leveraging the OpenAI Whisper API via Openclaw Skills.
Obsidian Vault Automation
A powerful integration for managing Obsidian vaults, enabling automated note creation, searching, and link-safe refactoring via the command line.
Things 3 CLI for macOS
A powerful macOS integration that enables AI agents to read from and write to the Things 3 task manager database and URL scheme.
wacli
A robust command-line interface for sending WhatsApp messages, managing files, and searching chat history via AI agents.
Featured*
