ask-search
llms.txt540 starsChanged 3 days agoMITView on GitHub
ask-search is a self-hosted web search CLI and MCP server for AI agents. It wraps SearxNG so Claude Code, OpenClaw, Antigravity or any shell can search Google, Bing, DuckDuckGo, Brave and more without a paid search API key. Tavily is an optional backend. 给 AI Agent 用的自托管联网搜索:包装 SearxNG,一条命令或一个 MCP server 即可聚合多个搜索引擎,无需付费搜索 API key。 A small Python CLI (ask-search) plus an MCP server. For people running AI agents that need web search and prefer not to depend on a paid search…
- Reads credentials
- Installs packages
In this repositoryOne other file it gives its agents.
- ask-searchSkill
The file
Files
# ask-search > ask-search is a self-hosted web search CLI and MCP server for AI agents. It wraps SearxNG so Claude Code, OpenClaw, Antigravity or any shell can search Google, Bing, DuckDuckGo, Brave and more without a paid search API key. Tavily is an optional backend. 给 AI Agent 用的自托管联网搜索:包装 SearxNG,一条命令或一个 MCP server 即可聚合多个搜索引擎,无需付费搜索 API key。 ## What it is / Who it's for A small Python CLI (`ask-search`) plus an MCP server. For people running AI agents that need web search and prefer not to depend on a paid search API. ## Install 1. Start SearxNG: `cd searxng && cp .env.example .env && docker compose up -d` 2. `bash install.sh` (or `sudo bash install.sh`, or `bash install.sh ~/.local/bin`) 3. `ask-search "hello world"` ## CLI usage `ask-search "query" [options]` | Flag | Short | Description | |------|-------|-------------| | `--num N` | `-n` | Max results (default 10) | | `--engines` | `-e` | e.g. google,bing,duckduckgo,brave | | `--lang` | `-l` | e.g. zh-CN, en, ja, ko | | `--categories` | `-c` | e.g. general,news,images,science | | `--json` | `-j` | Raw JSON output | | `--urls-only` | `-u` | URLs only | | `--version` | `-V` | Print version | ## Environment variables - `SEARXNG_URL` — SearxNG endpoint (default `http://localhost:8080`) - `SEARCH_PROVIDER` — `searxng` (default) or `tavily` - `TAVILY_API_KEY` — required when `SEARCH_PROVIDER=tavily` ## MCP tools - `web_search` — search via SearxNG - `web_search_news` — news search via SearxNG - `web_search_tavily` — search via Tavily (needs `TAVILY_API_KEY`) The MCP server needs Python 3.10+ and `pip install mcp`. ## Limitations - Requires a SearxNG instance with `json` in `search.formats`. - In Tavily mode, `--engines`, `--lang` and `--categories` are ignored. - Fetching full pages (deep-dive) can be blocked by the target site's anti-bot measures. ## Links - [README](README.md) - [SKILL.md](SKILL.md) - [LICENSE](LICENSE) - Repository: https://github.com/ythx-101/ask-search
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

