Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Summarizes URLs or text using Kagi's Universal Summarizer API. Supports multiple engines including the enterprise-grade Muriel model, bullet-point takeaways, and output translation to 28 languages. Works with articles, PDFs, video transcripts, forum threads, and academic papers.
Uses Kagi's API directly with a prebuilt Go binary, so there are no runtime dependencies and startup is fast compared to browser-based or LLM-chained summarization tools.
When to use it
- Summarize a research paper from arXiv without reading the full text
- Get bullet-point takeaways from a long news article
- Summarize a document and translate the result to Japanese
- Quickly digest a forum thread or blog post via URL
- Extract key points from a PDF or video transcript
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: summarize https://arxiv.org/abs/1706.03762 with bullet points using the best engine
- 1Receives the URL and user preferences for bullet points and high quality
- 2Calls kagi-summarizer with --engine muriel and --type takeaway flags
- 3Sends the request to Kagi's Universal Summarizer API
- 4Prints bullet-point takeaways to stdout with token usage on stderr
Structured bullet-point summary of the Transformer paper with API balance and token count reported
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.