web-search

Fetch live web content via the fetch MCP server and cite sources.

Updated Jan 27, 2026
One-click install
npx skills add https://github.com/stefanbosak/claude-cli --skill web-search-stefanbosak
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: web-search
Source: https://github.com/stefanbosak/claude-cli/tree/main/.claude/skills/web-search
Command: npx skills add https://github.com/stefanbosak/claude-cli --skill web-search-stefanbosak

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Real-time web search to retrieve current information from the internet, enabling up-to-date answers when training data is outdated or incomplete.

Core Features & Use Cases

  • Live retrieval: Fetch content from public URLs and summarize key points.
  • Source citation: Provide sources with direct URLs for verification and reproducibility.
  • Cross-source synthesis: Compare and synthesize information from multiple live sources to improve accuracy.

Quick Start

Fetch the target URL with the fetch MCP server and provide a concise summary.

Frequently Asked Questions about web-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I fetch live web content to answer questions about current events?

To fetch live web content for current events, this Skill retrieves real-time data from public URLs using the fetch MCP server, summarizes key points, and cites sources for verification.

Can I get up-to-date information on software releases when training data is outdated?

Yes, you can get up-to-date information on software releases by fetching live web content, which retrieves current changelogs and release notes directly from the internet to fill knowledge gaps.

What's the best way to cross-reference multiple web sources for accuracy?

The best way to cross-reference multiple sources is to fetch live web content from several URLs simultaneously, then compare and synthesize the retrieved markdown data to improve answer accuracy.

Do I need to configure an MCP server to retrieve real-time web data?

Yes, you need the fetch MCP server configured in the .claude/mcp-config.json file to retrieve real-time web content, request markdown output for parsing, and provide cited source URLs.

Does fetching live web content always include source citations?

Yes, fetching live web content always includes source citations with direct URLs, ensuring that retrieved information from current events or software releases remains verifiable and reproducible.

Why use live web search instead of relying on existing training data?

Use live web search when training data is outdated or incomplete, as fetching current web content provides fresh information for recent software releases, changelogs, and ongoing current events.