← 返回 Skills 市场
1132
总下载
4
收藏
11
当前安装
5
版本数
在 OpenClaw 中安装
/install tavilysearch
功能描述
Tavily 搜索引擎集成,支持 web 搜索、内容提取、实时新闻查询等功能。使用场景:需要搜索最新网络信息、查找实时新闻、获取专业资料、验证事实准确性等。
安全使用建议
This skill is coherent with its description: it will read a TAVILY_API_KEY from a .env file in the skill folder and make HTTPS calls to api.tavily.com, consuming your Tavily credits. Before installing: (1) Verify you trust the Tavily service and the skill source (no homepage or known owner listed). (2) Keep only the API key in the .env (don’t put other secrets there), and use a limited-scope API key if Tavily supports it. (3) Be aware the script disables some urllib3/TLS warnings — this may hide TLS/certificate issues, so confirm network calls go to the legitimate Tavily domain. (4) Run pip installs in an isolated environment (venv) and monitor API usage/credit consumption. If you want higher assurance, inspect the file scripts/search.py locally (it’s small and readable) or run it in an isolated container with a rotated API key.
功能分析
Type: OpenClaw Skill
Name: tavilysearch
Version: 1.0.4
The skill bundle provides a standard and well-documented integration for the Tavily Search API. The Python script `scripts/search.py` implements core functionalities like web searching, content extraction, and site crawling by making authorized HTTP requests to the official `api.tavily.com` endpoints. There is no evidence of data exfiltration, malicious execution, or prompt injection; the code specifically limits environment variable access to the required `TAVILY_API_KEY` and lacks any obfuscation or unauthorized network activity.
能力评估
Purpose & Capability
Name/description (web search, extraction, crawl, research) align with the included code and docs. The only external network calls are to https://api.tavily.com and the declared env var TAVILY_API_KEY is the expected credential for that service.
Instruction Scope
SKILL.md instructs installing requirements, creating a .env with TAVILY_API_KEY, and calling the included scripts. The runtime instructions do not ask for unrelated files, other credentials, or system-wide data. The code reads the skill-directory .env and uses only the API key.
Install Mechanism
No install spec in registry; SKILL.md asks users to run 'pip install -r requirements.txt' (requests, urllib3) which is proportional and expected. Nothing is downloaded from arbitrary URLs or installed from untrusted sources.
Credentials
Only TAVILY_API_KEY is required (declared in SKILL.md). The code does not request other secrets or unrelated environment variables or config paths.
Persistence & Privilege
Skill does not request 'always: true', does not modify other skills, and has no privileged system persistence. It runs on-demand and only uses the API key provided.
如何使用
- 确保已安装 OpenClaw(本地或 Docker 部署)
- 在对话框中输入安装命令:
/install tavilysearch - 安装完成后,直接呼叫该 Skill 的名称或使用
/tavilysearch触发 - 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
v1.0.4
v1.0.4 增强兼容性与依赖声明:
- 修复环境变量读取逻辑,仅读取 TAVILY_API_KEY,提升安全性
- 在 SKILL.md 中添加环境变量注册声明,确保需求一致
- 移除未使用的 python-dotenv 依赖,优化依赖管理
- 新增 urllib3 依赖,保证脚本正常运行
- 所有核心功能测试通过
v1.0.3
Tavily Search Skill v1.0.1
- 修复环境变量读取逻辑,仅读取 TAVILY_API_KEY,避免暴露敏感信息
- 在 SKILL.md 中补充环境变量声明,技能注册更精准
- 移除未使用的 python-dotenv 依赖,简化依赖管理
- 所有核心功能测试通过
v1.0.2
No user-visible changes in this release (no file changes detected).
v1.0.1
- 修复环境变量读取逻辑,仅读取 TAVILY_API_KEY,提升安全性
- 在 SKILL.md 中新增环境变量声明,确保注册信息与实际依赖一致
- 保持所有核心功能的测试通过
v1.0.0
Tavily Search Skill v1.0.0
- 修复整站爬取参数命名错误(root_url → url)
- 修复深度研究相关参数(topic → input,citation_style → citation_format)
- 修正内容提取格式化,支持列表类型结果输出
- 新增开发者文档 references/api-docs.md
- 所有核心功能通过测试
元数据
常见问题
Tavily Search 是什么?
Tavily 搜索引擎集成,支持 web 搜索、内容提取、实时新闻查询等功能。使用场景:需要搜索最新网络信息、查找实时新闻、获取专业资料、验证事实准确性等。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 1132 次。
如何安装 Tavily Search?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install tavilysearch」即可一键安装,无需额外配置。
Tavily Search 是免费的吗?
是的,Tavily Search 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
Tavily Search 支持哪些平台?
Tavily 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 Tavily Search?
由 xRay(@raydoomed)开发并维护,当前版本 v1.0.4。
推荐 Skills