This gives Claude direct access to DuckDuckGo search results through the ddgr CLI tool, which is honestly one of the most straightforward ways to get current web data into your context. It returns clean JSON output you can parse immediately, supports time filters for recent results, and handles DuckDuckGo's Bang syntax for searching specific sites like Wikipedia or StackOverflow. The documentation is thorough about the required flags (always use --noua and --json) and covers practical scenarios like site-specific searches and instant answers. If you need Claude to look up anything beyond its training data cutoff or search documentation in real time, this is the tool to reach for.
- skills.sh installs
- 395
- GitHub stars
- 3
- last updated
- Aug 21, 2026
Install
$ npx -y skills add ysm-dev/ddgr-skill --skill ddgr --agent claude-codeInstalls into .claude/skills of the current project.
Files
ddgr Web Search
Search DuckDuckGo from the terminal using ddgr.
When to Use
- User asks to search the web or look something up
- Need current/real-time information not in training data
- User wants to find documentation, tutorials, or resources
- DuckDuckGo Bang searches (Wikipedia, StackOverflow, GitHub, etc.)
Prerequisites
ddgr must be installed: brew install ddgr or pip install ddgr
Basic Usage
ddgr --noua --np --json "search query"
Required flags:
--noua: Disable user agent (always use this)--np(noprompt): Exit after showing results, no interactive prompt--json: Output in JSON format for parsing (implies --np)
Options
Sponsors
Details
- Repository
- github.com/ysm-dev/ddgr-skill
- Category
- CLI & Terminal
- First seen
- Jun 3, 2026
