Duckducksearch

This wraps the duckduckgo-search Python package so you can query DuckDuckGo from Claude. You get text, image, video, and news search with filters for region, time range, safe search, and image properties like size or license type. The real draw is avoiding rate limits with proxy support including Tor, plus search operators for exact matches and file types. Results come back as structured dictionaries with URLs, titles, and metadata. Works as both a library and CLI tool if you need batch searches. Fair warning, it's explicitly educational use only and not meant for commercial scraping at scale.

skills.sh installs: 155
last updated: Jun 3, 2026

View on GitHub

Install

$ npx -y skills add xsailor511/duckduck-web-search --skill duckducksearch --agent claude-code

Installs into .claude/skills of the current project.

Files

DuckDuckGo Search Skill

Overview

This skill provides web search capabilities using the DuckDuckGo search engine via the duckduckgo-search Python package. It enables text, image, video, and news searches with various filtering options.

Installation

pip install -U duckduckgo_search

Core Functionality

Text Search

from duckduckgo_search import DDGS

# Basic text search
results = DDGS().text("python programming", max_results=10)

# Advanced text search with filters
results = DDGS().text(
    keywords="python programming",
    region="us-en",           # Region: us-en, uk-en, cn-zh, etc.
    safesearch="moderate",    # on, moderate, off
    timelimit="m",           # d (day), w (week), m (month), y (year)
    backend="auto",          # auto, html, lite, bing
    max_results=10
)