← 返回 Skills 市场
sexychina

Ask Search

作者 sexychina · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
343
总下载
0
收藏
3
当前安装
1
版本数
在 OpenClaw 中安装
/install ask-search
功能描述
Web search via self-hosted SearxNG. Aggregates Google, Bing, DuckDuckGo, Brave. Returns title/url/snippet. Zero API key required.
使用说明 (SKILL.md)

ask-search

Web search powered by SearxNG. Aggregates multiple search engines, zero API key, full privacy.

Usage

ask-search "your query"                    # top 10 results
ask-search "query" --num 5                 # limit results
ask-search "AI news" --categories news     # news only
ask-search "query" --lang zh-CN            # Chinese results
ask-search "query" --urls-only             # URL list (pipe to web_fetch)
ask-search "query" --json                  # raw JSON

Agent Workflow

  1. Run ask-search "topic" to get candidates
  2. Check snippet — if enough, answer directly
  3. If snippet truncated, use web_fetch on the URL for full content

Parameters

Flag Short Description
--num N -n Max results (default 10)
--engines -e google,bing,duckduckgo,brave
--lang -l zh-CN, en, ja, ko
--categories -c general,news,images,science
--json -j Raw JSON output
--urls-only -u URLs only

Setup

Requires SearxNG running locally. Set SEARXNG_URL if not on default port 8080.

安全使用建议
This skill appears to do what it says: it expects you to run a local SearxNG instance (default http://localhost:8080) and provides a CLI and optional MCP server integration. Before installing: (1) review and run install.sh as a non-root user or pass a user-writable install path instead of blindly using /usr/local/bin, (2) ensure you trust the repository source you cloned from, and (3) be aware that any 'deep-dive' fetching (web_fetch/curl/Playwright or SOCKS proxy instructions) will make outbound requests to the target websites or route traffic through your proxy — those network actions can expose queries or require you to trust remote endpoints. If you don't want to run a local SearxNG, do not set SEARXNG_URL to a remote service you don't control.
功能分析
Type: OpenClaw Skill Name: ask-search Version: 1.0.0 The ask-search skill is a legitimate tool designed to provide AI agents with web search capabilities via a self-hosted SearxNG instance. The codebase consists of a Python core script (scripts/core.py), an MCP server for integration with tools like Claude Code (mcp/server.py), and a standard installation script (install.sh). It uses subprocess.run with curl to perform network requests, but properly sanitizes query parameters using urllib.parse.urlencode to prevent injection. No evidence of data exfiltration, persistence mechanisms, or malicious prompt injection was found.
能力评估
Purpose & Capability
Name/description claim a local SearxNG search wrapper; included files (core.py, mcp/server.py, install.sh) implement that exact functionality and only depend on curl/python and an optional SEARXNG_URL. Nothing unrelated (cloud creds, extraneous services, or system-level access) is requested.
Instruction Scope
SKILL.md and README instruct the agent to run the CLI to get snippets and optionally call an external web_fetch for deep-dive content. Guidance on proxies, Playwright, and systemd is advisory for fetching blocked sites and is within scope for improving fetch reliability, not hidden data collection.
Install Mechanism
No remote arbitrary downloads or extract steps in the provided install.sh — it writes a small wrapper to the chosen bin dir and tests connectivity. README suggests git-cloning from GitHub (standard). No suspicious external URLs, shorteners, or binary drops are present in the installer.
Credentials
The skill requires no secrets or credentials. It optionally reads SEARXNG_URL (documented) which is proportionate to the purpose. There are no requests for unrelated env vars or config paths.
Persistence & Privilege
Skill is not always-enabled and does not attempt to change other skills or system-wide agent configs. The installer places a wrapper in a bin directory (normal) but does not persist credentials or enable autonomous escalation.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install ask-search
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /ask-search 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release of ask-search: simple, private web search via SearxNG. - Aggregates results from Google, Bing, DuckDuckGo, and Brave with no API key required. - Offers multiple filters (engine, language, category, number of results). - Provides several output formats: concise list, URLs only, or raw JSON. - Designed for agent workflows: fetch snippets quickly, escalate to web_fetch for full content if needed. - Simple local setup with customizable SearxNG URL.
元数据
Slug ask-search
版本 1.0.0
许可证 MIT-0
累计安装 5
当前安装数 3
历史版本数 1
常见问题

Ask Search 是什么?

Web search via self-hosted SearxNG. Aggregates Google, Bing, DuckDuckGo, Brave. Returns title/url/snippet. Zero API key required. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 343 次。

如何安装 Ask Search?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install ask-search」即可一键安装,无需额外配置。

Ask Search 是免费的吗?

是的,Ask Search 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Ask Search 支持哪些平台?

Ask Search 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Ask Search?

由 sexychina(@sexychina)开发并维护,当前版本 v1.0.0。

💬 留言讨论