← Back to Skills Marketplace
azzazhang

Bing Search (Free)

by azzazhang · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ suspicious
622
Downloads
0
Stars
6
Active Installs
1
Versions
Install in OpenClaw
/install bing-search-free
Description
免费网页搜索,使用 Bing + Jina.ai 提取内容摘要。无需 API Key,直接可用。适用于快速获取搜索结果和网页内容。
README (SKILL.md)

Bing Search Skill

免费网页搜索方案,基于 Bing 搜索 + Jina.ai 内容提取。

特性

  • 完全免费 — 无需 API Key
  • 🌐 Bing 搜索 — 全球搜索结果
  • 📄 内容摘要 — Jina.ai 自动提取页面核心内容
  • 快速响应 — 通常 1 秒内完成

使用方法

命令行

./scripts/bing_search.sh '\x3Cjson>'

示例:

# 基本搜索
./scripts/bing_search.sh '{"query": "Python 教程"}'

# 指定结果数量
./scripts/bing_search.sh '{"query": "AI news", "max_results": 5}'

# 中文搜索
./scripts/bing_search.sh '{"query": "人工智能 最新消息", "max_results": 3}'

参数说明

参数 类型 默认值 说明
query string 必填 搜索关键词
max_results number 5 返回结果数量 (1-10)

输出格式

{
  "query": "搜索关键词",
  "results": [
    {
      "url": "https://example.com",
      "title": "页面标题",
      "content": "页面内容摘要...",
      "score": 0.95
    }
  ],
  "response_time": 0.75
}

技术原理

  1. Bing 搜索 — 获取搜索结果列表
  2. Jina.ai 提取https://r.jina.ai/http://bing.com/search?q=关键词
  3. 结果解析 — 提取标题、URL、内容摘要

限制

  • 免费服务,有请求频率限制
  • 不适合大规模自动化搜索
  • 结果质量依赖 Bing 和 Jina.ai

故障排除

问题: 返回空结果 解决: 检查网络连接,稍后重试

问题: 请求频繁被拒 解决: 等待几秒后再试,或减少请求频率


来源: agents-skills-personal → openclaw-workspace

Usage Guidance
What to consider before installing/use: - Functionally consistent: the script does what the description claims — it fetches Bing results via the Jina.ai proxy (r.jina.ai) and extracts summaries. However, the script requires common CLI tools (curl, jq, python3, bc, grep) even though the skill metadata lists none. Ensure these are available. - Privacy: requests and page content go through r.jina.ai, a third-party service. If you search sensitive URLs or content, that content may be seen or logged by Jina.ai. Avoid using this for confidential data. - Security and robustness: the Python parsing embeds the raw HTTP response into an inline -c string; that is brittle and could break or be abused by specially crafted responses. Consider running the script in a sandbox, or inspect/modify the script to write RESPONSE to a temporary file and have Python read the file (safe quoting), or use a safer parser. - Recommended safeguards: run in an isolated environment, verify/replace the parsing routine to avoid injecting untrusted content into code, add input validation and rate-limiting, and test behavior on varied responses (CAPTCHA, HTML edge cases). If you need production-grade search, prefer official APIs with documented SLAs and privacy terms. - If you want me to: I can (a) produce a safer patched version of the script that avoids embedding raw responses into python -c and declares required binaries, or (b) perform a line-by-line explanation of the current script and the exact commands you should run to sandbox it.
Capability Analysis
Type: OpenClaw Skill Name: bing-search-free Version: 1.0.0 The skill provides a Bing search interface by proxying requests through the Jina.ai Reader service. While the functionality aligns with its stated purpose, scripts/bing_search.sh contains a significant security vulnerability: it unsafely injects the raw search response into an inline Python script using triple quotes ('''$RESPONSE'''). This allows for potential Remote Code Execution (RCE) if a search result contains specifically crafted content designed to escape the Python string literal and execute arbitrary code.
Capability Assessment
Purpose & Capability
The skill's code and SKILL.md implement Bing search via Jina.ai as claimed. However the package metadata says no required binaries, while the included script actually depends on curl, jq, python3, bc, grep and standard shell utilities. Declaring no required binaries is inconsistent and could mislead users about runtime needs.
Instruction Scope
SKILL.md limits runtime behavior to running scripts/bing_search.sh, which only queries https://r.jina.ai/http://bing.com/search and parses results. That stays within the stated purpose, but the script embeds the raw HTTP response directly into a python3 -c string for parsing (response = '''$RESPONSE'''). This is brittle and may break or be abused if the response contains sequences that close the string; it presents a parsing/injection risk. The script also lacks robust sanitization and error handling for edge cases.
Install Mechanism
No install spec—this is instruction-only plus an included script. That is the lowest install risk (nothing downloaded at install time).
Credentials
The skill requests no credentials or environment variables, which is appropriate for a public free-search wrapper. Note: it does send queries and retrieved page content to r.jina.ai (third-party service), which has privacy implications not represented as credential/config requirements.
Persistence & Privilege
always is false, the skill does not request elevated or persistent privileges and does not modify other skills or system-wide settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install bing-search-free
  3. After installation, invoke the skill by name or use /bing-search-free
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release - 免费网页搜索方案
Metadata
Slug bing-search-free
Version 1.0.0
License MIT-0
All-time Installs 6
Active Installs 6
Total Versions 1
Frequently Asked Questions

What is Bing Search (Free)?

免费网页搜索,使用 Bing + Jina.ai 提取内容摘要。无需 API Key,直接可用。适用于快速获取搜索结果和网页内容。 It is an AI Agent Skill for Claude Code / OpenClaw, with 622 downloads so far.

How do I install Bing Search (Free)?

Run "/install bing-search-free" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Bing Search (Free) free?

Yes, Bing Search (Free) is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does Bing Search (Free) support?

Bing Search (Free) is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Bing Search (Free)?

It is built and maintained by azzazhang (@azzazhang); the current version is v1.0.0.

💬 Comments