skills.rest
dyoshikawa
dyoshikawaCOMMUNITY
Explore all skills by this author →
GitHub

webseek

Execute web searches across Google, OpenAI, and Gemini providers.

1|Updated Jun 25, 2026
Search Tags:#automation#cli#web search#mcp#google search#gemini#openai
One-click install
npx skills add https://github.com/dyoshikawa/webseek --skill webseek
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: webseek
Source: https://github.com/dyoshikawa/webseek/tree/main/skills/webseek
Command: npx skills add https://github.com/dyoshikawa/webseek --skill webseek

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill solves the fragmentation of web search tools by providing a single, unified interface to access multiple search providers, including Google, OpenAI, and Gemini, through a consistent CLI, MCP server, or programmatic API.

Core Features & Use Cases

  • Multi-Provider Support: Seamlessly switch between Google Custom Search, OpenAI web search, and Gemini grounding.
  • Flexible Deployment: Run as a standalone CLI tool, an MCP server for AI agents, or integrate directly into your own TypeScript/JavaScript applications.
  • Use Case: Use this tool to quickly gather research data from multiple search engines into a normalized JSON format for automated analysis or to provide real-time web access to an LLM-based agent.

Quick Start

Use the webseek skill to perform a search for the latest developments in artificial intelligence using the google provider.

Frequently Asked Questions about webseek

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

FAQPage Schema
How do I execute web searches across multiple providers like Google and OpenAI using a single interface?▼

You can execute web searches across multiple providers by using a unified interface that supports Google, OpenAI, and Gemini. This interface normalizes the search results into a consistent JSON format for automated analysis.

What is the best way to give an MCP-compatible AI agent access to real-time web search?▼

Providing real-time web search to an MCP-compatible AI agent is best done by running a unified search interface as an MCP server. This allows the agent to directly query providers like Google and Gemini for information retrieval.

Do I need API keys to use Google, OpenAI, and Gemini for web search?▼

Yes, you need API keys to use Google, OpenAI, and Gemini for web search. Valid API keys for your selected search provider must be configured in the environment variables before executing queries.

Can I use a web search tool programmatically within my TypeScript or JavaScript applications?▼

Yes, you can use a multi-provider web search tool programmatically within TypeScript or JavaScript applications. It provides a programmatic API that allows direct integration to gather research data.

Does a unified web search interface work as a standalone command line tool?▼

Yes, a unified web search interface works as a standalone command line tool. It offers flexible deployment options, allowing CLI users to perform searches directly from the terminal without needing a separate server.

Related Skills

Explore More Software Engineering Skills→
buddypiabuddypia

deep-research

Run long-running research tasks via OpenAI and Google Gemini Deep Research APIs.

Community
Advanced
Blake-JohnBlake-John

pymupdf

Convert PDF documents into Markdown, JSON, and text for LLM workflows.

Community
Intermediate
olostepolostep

search

Search the web in real time using Olostep tools for answers with citations.

Community
Intermediate
juanrejuanre
3

providers

Switch between LLM providers and access raw SDKs via get_provider().

Community
Intermediate