What are the requirements to use this skill?

It requires internet access and the installation of the parallel-cli tool, which can be installed via a simple shell script or through pipx.

Can I save the results of my search?

Yes, the skill automatically outputs full search results to a JSON file on your local system, allowing you to reference the raw data in future prompts.

How does it handle citations?

The skill follows a strict protocol where every claim must have an inline markdown link to the source URL, and a full Sources section is appended to every response.

When should I use this instead of parallel-deep-research?

Use Parallel Web Search for standard lookups, news, and general investigations. Only switch to deep research when you specifically require an exhaustive, multi-step analysis.

What is Parallel Web Search?

Parallel Web Search is a specialized skill for Claude Code that enables the AI to perform internet searches, gather real-time data, and provide cited information for any research query.

Parallel Web Search

Name: Parallel Web Search
Author: parallel-web

byparallel-web

•

ウェブスクレイピングとデータ収集

Conducts fast, cost-effective web research and information lookups with automated inline citations and structured data output.

Parallel Web Search is the primary capability for performing real-time internet research, investigations, and data lookups within Claude Code. It utilizes the parallel-cli to execute broad objective-based searches and targeted keyword queries, extracting specific facts, dates, and metrics from web excerpts. Designed for accuracy and transparency, the skill enforces a strict citation policy for every claim made and automatically saves full search results to a local JSON file for persistent access and follow-up analysis.

主な機能

01Optimized for cost-effectiveness compared to multi-step deep research tools.

02Enforces mandatory inline citations and a structured 'Sources' section for all findings.

03Supports advanced filtering by date ranges and specific domain inclusions.

0413 GitHub stars

05Exports full search results to local JSON files for persistent follow-up and deeper analysis.

06Executes high-speed web searches using natural language objectives and keyword flags.

ユースケース

01Fact-checking claims and gathering verifiable statistics with direct source links.

02Performing competitive analysis or technology comparisons using real-time web data.

03Researching current software documentation, library versions, or breaking tech news.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add parallel-web/parallel-agent-skills parallel-web-search

For use in Claude.ai and ChatGPT

Download Skill