kagi-summarizer

What This Skill Does

The kagi-summarizer skill acts as a powerful analytical bridge between your workspace and Kagi’s Universal Summarizer API. It processes complex information from various sources including live URLs, academic papers, lengthy PDFs, and raw text snippets. By leveraging advanced summarization engines like Cecil, Agnes, and the enterprise-grade Muriel model, this skill distills verbose content into either concise prose or actionable bullet-point takeaways. It is an essential tool for knowledge workers who need to digest massive amounts of information rapidly without losing core context. Beyond simple summarization, the skill supports on-the-fly translation into 28 languages and offers native JSON output, making it highly compatible with automated data pipelines and AI-driven research workflows.

Installation

To integrate this skill, ensure you have your Kagi API credentials ready. First, install the binary via the OpenClaw command: clawhub install openclaw/skills/skills/joelazar/kagi-summarizer. Once installed, export your API key as an environment variable by adding export KAGI_API_KEY="your-api-key-here" to your shell configuration file (e.g., ~/.zprofile). The tool is designed for zero-dependency execution, ensuring that it remains lightweight and fast regardless of your environment.

Use Cases

This skill is perfect for researchers, students, and professionals who frequently navigate dense documentation. Use it to generate quick summaries of long technical blog posts, extract key findings from ArXiv papers, or translate foreign-language forum threads into your native tongue. It is also an excellent utility for preparing briefing documents or summarizing video transcripts when you have limited time to watch full-length presentations.

Example Prompts

• "Summarize this article for me: https://example.com/deep-dive-into-llms and give me the key points as bullets."
• "Using the Muriel engine, summarize the text I've piped into the system and translate the results into Japanese."
• "What are the main arguments in this paper? https://arxiv.org/abs/1706.03762. Please provide a brief summary in JSON format."

Tips & Limitations

• Cost Efficiency: Keep in mind that summarization is billed per 1,000 tokens. Utilize caching where possible to avoid redundant API charges; the system handles cached requests for free.
• Engine Selection: While 'Cecil' is the default and balanced choice, switch to 'Muriel' when dealing with highly complex academic or technical material requiring deep nuance.
• Timeouts: The default timeout is 120 seconds, which is usually sufficient for most web pages. If you are processing exceptionally large documents, consider adjusting the timeout flag to prevent premature connection termination.