web-search

Iteratively search the open web and synthesize findings with inline citations.

Updated Mar 22, 2026
One-click install
npx skills add https://github.com/shusuke-ioku/r2 --skill web-search-shusuke-ioku
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: web-search
Source: https://github.com/shusuke-ioku/r2/tree/main/src/r2/template/.claude/skills/web-search
Command: npx skills add https://github.com/shusuke-ioku/r2 --skill web-search-shusuke-ioku

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Open-web information gathering is often shallow and manual; this skill orchestrates iterative search to collect deeper open-web sources and produce a structured synthesis with citations.

Core Features & Use Cases

  • Iterative open-web research using Codex CLI for broad exploration and Claude Code's WebSearch/WebFetch for quick lookups.
  • Multi-step refinement: query, read results, identify gaps, refine, and fetch full pages as needed.
  • Use Case: Build a background context for a policy analysis with diverse, verifiable sources beyond academic databases.

Quick Start

Initiate an iterative open-web search to gather background information and synthesize findings with citations.

Frequently Asked Questions about web-search

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

FAQPage Schema
How do I compile background information from open-web sources with diverse perspectives?

Iterative web research compiles open-web sources by querying, reading results, identifying gaps, refining queries, and fetching full pages to produce a structured synthesis with inline citations.

What is the best way to fact-check current events and build context beyond academic databases?

Open-web iterative research fact-checks current events by orchestrating broad exploration and quick lookups, delivering structured context-building with verifiable sources and a final sources list.

How does iterative web research with Codex CLI and Claude Code work for source synthesis?

Iterative web research leverages Codex CLI for heavy-lifting broad exploration and Claude Code's WebSearch/WebFetch for quick lookups, applying multi-step refinement to synthesize sources with inline citations.

Can I use this iterative research approach for policy analysis requiring verifiable sources?

Iterative web research suits policy analysis by collecting diverse, verifiable open-web sources beyond academic databases, applying multi-step query refinement to build background context with structured citations.

When do I need multi-step refinement for open-web research instead of a single search?

Multi-step refinement is needed when initial search results reveal gaps, requiring query refinement and full page fetches to compile deeper open-web sources and deliver a complete structured synthesis.

Does iterative web research provide inline citations and a final sources list?

Iterative web research delivers a structured synthesis with inline citations throughout the compiled background information and a final sources list for verifiable open-web references.