skills.rest
mammut001
mammut001COMMUNITY
Explore all skills by this author →
GitHub

web_research

Navigate live web pages and extract cited evidence for research questions.

2|Updated Mar 13, 2026
Search Tags:#research#evidence#citation#browser#web#live-data
One-click install
npx skills add https://github.com/mammut001/pipi-shrimp-agent --skill web-research-mammut001
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: web_research
Source: https://github.com/mammut001/pipi-shrimp-agent/tree/main/src/skills/web_research
Command: npx skills add https://github.com/mammut001/pipi-shrimp-agent --skill web-research-mammut001

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Research a topic in a live browser using PageState-based navigation, extraction, and evidence synthesis.

Core Features & Use Cases

  • Live web exploration: Navigate to target pages and perform on-demand information gathering.
  • Structured extraction & evidence synthesis: Collect content from page state outputs and compile grounded conclusions with citations.
  • Use Case: If you need current specs from competing products, compare them and summarize with sources.

Quick Start

Research a topic live on the web and present a concise, evidence-backed answer with sources.

Frequently Asked Questions about web_research

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

FAQPage Schema
How do I research live web data and get evidence-backed answers?▼

To research live web data, you navigate live pages using a PageState-based workflow to extract content and compile grounded conclusions with citations. This enables real-time information gathering and evidence synthesis directly from target sources.

Can I compare product specs across multiple live web sources?▼

Yes, you can compare product specs across sources by navigating to target pages, extracting structured content from page state outputs, and synthesizing the data into a concise comparison with citations for each spec.

What is the best way to summarize current events with citations from the web?▼

The best way to summarize current events with citations is using live web exploration to gather real-time data, then compiling the extracted page content into a synthesized summary that explicitly references the source pages.

Does this web research approach support browser navigation and page inspection?▼

Yes, this approach supports browser navigation, page-state inspection, click and type actions, and content extraction. These actions are required to interact with live pages and gather evidence-backed information.

How does a PageState-based workflow extract content for research?▼

A PageState-based workflow extracts content by inspecting the state of a live page after navigation or input actions, collecting the rendered content, and compiling it into evidence-backed conclusions with source citations.

Are there limitations when extracting live data for current-events questions?▼

Extraction of live data for current-events questions depends on navigating target pages and inspecting page states, so limitations arise if pages block browser access or require complex interactions beyond standard click and type actions.

Related Skills

Explore More Education & Research Skills→
tiankong0101-bytetiankong0101-byte

agent-browser

Automate browser-based web interaction via a WebSocket endpoint.

Community
Intermediate
Formspec-orgFormspec-org
4

semi-formal-code-review

Review code changes with file:line evidence and test relevance.

Official
Advanced
joeseesunjoeseesun
978

qiaomu-opencli-browser

Automate browser interactions via existing Chrome sessions with structured commands.

Community
Intermediate
richardnguyen0715richardnguyen0715

scrapling

Scrape static and JavaScript-rendered web content with stealth and Cloudflare bypass.

Community
Advanced