Turn your AI coding assistant into a web scraping expert
The Nimble Web Expert skill gives your agent the ability to search, extract, scrape, map, crawl, and run pre-built extraction agents across any website. Fast, incremental, and immediately responsive. This is the only way your AI assistant can access live websites.Powered by the Nimble CLI and built on the open-source Agent Skills standard for cross-platform agent compatibility.
Accurate, real-time web search. 8 focus modes: general, coding, news, academic, shopping, social, geo, location.
Extract
Scalable data collection with stealth unblocking. Get clean, real-time HTML and structured data from any URL. Supports JS rendering and browser emulation.
Map
Fast URL discovery and site structure mapping. Plan extraction workflows before running them.
Crawl
Extract contents from entire websites in a single request. Collect large volumes of web data automatically.
Agents
Run pre-built extraction agents for structured data from popular websites (Amazon, Google, LinkedIn, and hundreds more).
Installs the Nimble skills and configures the MCP server.
claude plugin marketplace add Nimbleway/agent-skills && \claude plugin install nimble@nimble-plugin-marketplace
2
Verify installation
Run /mcp and confirm nimble appears in your connected servers.
1
Add the MCP server
One click, then confirm the install in Cursor.Then replace NIMBLE_API_KEY in Settings → MCP Servers → nimble-mcp-server with your Nimble API key. The deeplink ships a placeholder, not a key.
2
Add the skills
Teaches the agent which Nimble product fits a given job.
npx skills add Nimbleway/agent-skills -a cursor
3
Verify installation
Open Settings, then MCP, and confirm nimble-mcp-server is listed.
1
Run this command in your terminal
Installs the Nimble skills into your project.
npx skills add Nimbleway/agent-skills
2
Verify installation
Run npx skills add Nimbleway/agent-skills --list to see what was installed.
The Nimble Web Expert skill activates automatically when you ask for live web data:
"Search for recent AI developments" -> nimble search --query "recent AI developments" --search-depth lite"Extract the content from this URL" -> nimble extract run --url "https://example.com" --parse --format markdown"Scrape prices from this product page" -> nimble extract run --url "https://example.com/product" --render --parse --format markdown"Map all the pages on this docs site" -> nimble map --url "https://docs.example.com" --limit 100"Crawl the API docs section" -> nimble crawl run --url "https://docs.example.com/api" --limit 50
Use this skill for immediate, one-off data needs. For workflows you will run
repeatedly, have it build an Extract Template
or schedule an Extract Job instead.
Accurate, real-time web search with 8 focus modes (general, coding, news, academic, shopping, social, geo, location).
# Lite search (default, fast, token-efficient)nimble search --query "React hooks tutorial" --focus coding --search-depth lite# Search with AI-generated answer summarynimble search --query "what is WebAssembly" --include-answer --search-depth lite# News search with time filternimble search --query "AI developments" --focus news --time-range week --search-depth lite# Domain-filtered searchnimble search --query "auth best practices" \ --include-domain github.com \ --include-domain stackoverflow.com \ --search-depth lite
Use --search-depth lite (default) for fast responses (1-3s). Use
--search-depth deep when you need full page content for archiving or
full-text analysis.
Get clean, structured data from any URL with stealth unblocking and JS rendering.
# Standard extraction (always use --parse --format markdown)nimble extract run --url "https://example.com/article" --parse --format markdown# Render JavaScript for SPAs and dynamic contentnimble extract run --url "https://example.com/app" --render --parse --format markdown# Extract with geolocationnimble extract run --url "https://example.com" --country US --city "New York" --parse --format markdown
Extract content from entire websites asynchronously.
# Start a crawl (always set --limit)nimble crawl run --url "https://docs.example.com" --limit 50# Crawl with path filteringnimble crawl run --url "https://example.com" \ --include-path "/docs" \ --include-path "/api" \ --limit 100# Check crawl statusnimble crawl status --id "crawl-id"# List all crawlsnimble crawl list
For LLM-friendly output, prefer map + extract --parse --format markdown on
individual pages. Crawl returns raw HTML which can be very large.
This skill follows the Agent Skills open-source standard, making it compatible with multiple AI agent platforms. Install using the Skills CLI, the standard package manager for the Agent Skills ecosystem.
Claude Code
Cursor
Vercel CLI