Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of recent attention, adoption, and active maintenance.
Ask-search brings web search into the terminal via a self-hosted SearxNG instance that queries Google, Bing, DuckDuckGo, and Brave simultaneously. It gives privacy-conscious searchers and AI agent builders a way to retrieve web results without exposing a commercial API key or relying on a single engine. Output can be limited, filtered by language and category, or returned as raw JSON for programmatic use.
WHO IT'S FOR
Self-hosted / privacy-conscious users
run private web searches without API keys
AI agent / automation builders
integrate aggregated web search into agent workflows
Developers in multi-search-engine environments
compare results across Google, Bing, DuckDuckGo, Brave
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Compatibility not documented
Use this skill
Follow the documented setup, then try a first task.
Quick start with existing SearxNG
Installs: Clone the complete ask-search repository and install its global ask-search command; the repository includes the ask-search skill.
Before you start
SearxNG already running locally with JSON output enabled
Python 3.8 or newer
curl
git and bash
In a terminal, clone the repository, enter its directory, run the installer, and perform the documented test search.
git clone https://github.com/ythx-101/ask-search
cd ask-search
bash install.sh
ask-search "hello world"
Ensure the installed ask-search command can reach a running SearxNG instance; set SEARXNG_URL if it is not using the default local port 8080.
Use the ask-search skill to search the news category for [topic], limiting the search to five results. Report each result's title, URL, and snippet without fetching the full pages.
