native-web-search

Run native web searches and return concise findings with canonical URLs.

1|1|Updated Feb 5, 2026
One-click install
npx skills add https://github.com/ntk148v/skills --skill native-web-search-ntk148v
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: native-web-search
Source: https://github.com/ntk148v/skills/tree/main/skills/native-web-search
Command: npx skills add https://github.com/ntk148v/skills --skill native-web-search-ntk148v

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you quickly research a topic on the internet and produce a concise, purpose-driven summary with reliable, full canonical source URLs.

Core Features & Use Cases

  • Fast native web search: Runs a deterministic search flow using native web-search tools via supported providers.
  • Concise summaries for a specific purpose: Tailors findings to a stated reason (e.g., updating dependency notes or planning migration steps).
  • Full canonical URLs and source disagreement: Requires complete https:// links for key findings and highlights when sources conflict.
  • Provider/model flexibility: Supports openai-codex or anthropic, with options for model selection and timeouts.

Quick Start

Run node search.mjs with your query and a purpose, such as asking it to find the latest Python release details for updating dependency notes.

Frequently Asked Questions about native-web-search

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

FAQPage Schema
How do I get web search results with full canonical URLs for migration planning?

Web search with full canonical URLs is performed by running a native search flow that returns concise, purpose-driven summaries. This Skill aggregates findings via provider-specific APIs and highlights source disagreements for migration decision-making.

Can I use native web search tools to compare conflicting sources for dependency updates?

Native web search tools can compare conflicting sources by streaming results and highlighting disagreements. This Skill tailors findings to a specific purpose like dependency updates, returning complete https links for verification.

Does this web search summary tool support both OpenAI and Anthropic providers?

The web search summary tool supports both OpenAI Codex and Anthropic web search APIs. It offers provider and model selection flexibility, allowing you to run deterministic search flows and output results in text or JSON form.

What is the best way to run a fast internet research query and output JSON findings?

The best way to run fast internet research is by executing a script with your query and purpose, which triggers native web search APIs. The tool aggregates findings and outputs practical multi-finding summaries in either text or JSON format.

How does source disagreement awareness work in web research summaries?

Source disagreement awareness works by evaluating aggregated search results and explicitly highlighting when sources conflict. This ensures your research summary provides a balanced view for development and migration decision-making.

When do I need a purpose-driven web search summary instead of a standard search?

You need a purpose-driven web search summary when requiring evidence-backed findings for specific tasks like updating dependency notes or planning migration steps. It tailors concise results with full canonical URLs rather than returning generic links.