skills.rest
JaviMontano
JaviMontanoCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

web-research

Scrapes, crawls, searches, and synthesizes live web content via firecrawl with URL provenance.

Updated Jun 28, 2026
Search Tags:#data extraction#search#firecrawl#provenance#web scraping#deep research#crawl
One-click install
npx skills add https://github.com/JaviMontano/claude-plugins --skill web-research-javimontano
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: web-research
Source: https://github.com/JaviMontano/claude-plugins/tree/main/plugins/claude-native-toolkit/skills/web-research
Command: npx skills add https://github.com/JaviMontano/claude-plugins --skill web-research-javimontano

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes assets (resource) components.

What problem does it solve? Getting reliable, source-cited data from the live web requires choosing the right access mode (single page, whole site, search, or multi-source research) and enforcing provenance on every claim. This Skill governs live web access through firecrawl so every extracted fact carries its source URL. ## Core Features & Use Cases - Mode selection: Routes each request to the right firecrawl mode — scrape (one page), crawl (a site), search (the web), map (discover site URLs), or deep-research (multi-source synthesis). - Structured extraction: Parses scraped markdown into structured data, since no native URL-based extract mode exists. - Provenance discipline: Requires a source URL for every data point and flags uncited claims. - Use Case: Ask for the current state of WebGPU browser support; the Skill searches for primary sources, scrapes support matrices per browser, and returns a table with per-claim source URLs. ## Quick Start Use web-research to scrape https://example.com/pricing and list the plan names and prices with their source URLs.

Frequently Asked Questions about web-research

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

FAQPage Schema
How do I scrape a web page and get structured data?▼

Use the firecrawl scrape mode to fetch the page as markdown, then parse that markdown into structured fields. There is no native URL-based extract mode, so scraping plus parsing is the supported path.

What is the difference between firecrawl scrape, crawl, search, and map?▼

Scrape retrieves one page, crawl traverses an entire site, search queries the web, and map discovers the URLs a site contains. Deep research synthesizes findings across multiple sources.

When should I use NotebookLM instead of live web research?▼

Use notebooklm-research when synthesizing documents already loaded into a NotebookLM notebook. Live web research is for fetching fresh content from URLs, not for pre-loaded document collections.

Does web scraping require citing source URLs?▼

Yes. Every data point must carry its source URL as a provenance discipline, and requests to omit sources are refused. Uncited claims are flagged rather than presented as verified facts.

What happens when a page hits rate limits or heavy JavaScript?▼

The Skill degrades gracefully and reports what could not be covered. Rate-limited or JavaScript-heavy pages are edge cases where partial results are returned with gaps explicitly noted.

Related Skills

Explore More Education & Research Skills→
firecrawlfirecrawl
616

firecrawl

Search, scrape, crawl, and interact with websites via the Firecrawl CLI.

Official
Intermediate
dotlab-hqdotlab-hq

web-research

Coordinate web searches and synthesize findings into a cited research report.

Official
Intermediate
george11642george11642
2

mcp-notebooklm

Automate multi-source research and notebook synthesis with NotebookLM MCP.

Community
Advanced
DTMC-marketplaceDTMC-marketplace

gemini-mcp-integration

Configure MCP servers with Gemini CLI via HTTP, SSE, and Stdio transports.

Official
Intermediate