Real-time web search using Tavily API optimized for AI agents. Use when you need current information, recent news, factual research, competitive analysis, or...
数据来源:ClawHub。 在 ClawSkills 查看
选择你使用的 Agent
方法一:命令行安装(推荐)
推荐(无需提前安装 clawhub)
npx clawhub@latest --dir ~/.claude/skills install moochmaniac-tavily-search或使用 clawhub CLI(需提前安装)
clawhub --dir ~/.claude/skills install moochmaniac-tavily-search⚠️ 需要 Node.js 18+,没有 Node?请使用下方方法二直接下载 ZIP。 安装 Node.js →
方法二:手动下载安装(无需 Node)
下载 ZIP,解压后将文件夹放到以下路径,重启 Agent 即可:
安装路径
~/.claude/skills/moochmaniac-tavily-search/💡解压后将文件夹放到上方路径,重启 Agent 即可生效
--- name: tavily-search description: Real-time web search using Tavily API optimized for AI agents. Use when you need current information, recent news, factual research, competitive analysis, or any data not in your training cutoff. Searches return structured results with URLs, content snippets, relevance scores, and optional AI-generated answers. ---
Web search powered by Tavily API, optimized for AI agent consumption with structured, relevant results.
Basic search (5 results):
python3 scripts/search.py "your query here"
With options:
# Fewer results
python3 scripts/search.py "AI agents" --max-results 3
# AI-generated answer
python3 scripts/search.py "what is OpenClaw" --answer
# Advanced search (deeper crawling)
python3 scripts/search.py "casino bonuses" --depth advanced
# Include images
python3 scripts/search.py "WordPress themes" --images
# Raw JSON output
python3 scripts/search.py "n8n automation" --json
--max-results N: Number of results (1-10, default 5)--depth: Search depth - basic (default): Fast, good for most queries - advanced: Deeper crawling, more comprehensive
--answer: Include AI-generated answer summary--images: Include relevant image URLs--json: Output raw JSON instead of formatted textFormatted output includes:
- Title and URL - Relevance score (0-1) - Content snippet (truncated to 300 chars)
JSON output provides full Tavily API response for programmatic use.
The script looks for the Tavily API key in:
workspace/secrets/tavily_api_key (OpenClaw workspace)secrets/tavily_api_key (relative to current directory)TAVILY_API_KEY environment variableCurrent key is stored at: workspace/secrets/tavily_api_key
Research task:
python3 scripts/search.py "WordPress plugin security best practices 2026" --max-results 5 --answer
Quick fact check:
python3 scripts/search.py "OpenClaw GitHub repository" --max-results 3
Competitive analysis:
python3 scripts/search.py "top casino affiliate sites" --depth advanced --max-results 10
News monitoring:
python3 scripts/search.py "AI agent frameworks January 2026" --answer
--answer is useful for summarization tasks--depth advanced for comprehensive research (uses more API credits)--json output is useful for parsing and processing results programmatically安装 Tavily Web Search 后,可以对 AI 说这些话来触发它
Help me get started with Tavily Web Search
Explains what Tavily Web Search does, walks through the setup, and runs a quick demo based on your current project
Use Tavily Web Search to real-time web search using Tavily API optimized for AI agents
Invokes Tavily Web Search with the right parameters and returns the result directly in the conversation
What can I do with Tavily Web Search in my data & analytics workflow?
Lists the top use cases for Tavily Web Search, with example commands for each scenario
将技能文件夹放到 ~/.claude/skills/moochmaniac-tavily-search/ 目录(个人级,所有项目可用),或 .claude/skills/moochmaniac-tavily-search/(项目级)。重启 AI 客户端后,用 /moochmaniac-tavily-search 主动调用,或让 AI 根据上下文自动发现并使用。
Tavily Web Search 支持 Claude、Cursor、OpenClaw,可与这些 AI 平台无缝集成,扩展其能力。
Tavily Web Search 可免费安装使用。请查阅仓库了解许可证信息。
Real-time web search using Tavily API optimized for AI agents. Use when you need current information, recent news, factual research, competitive analysis, or...
Tavily Web Search 属于「Data & Analytics」分类,该分类的技能帮助 AI 智能体在此领域执行专业任务。
Automate my data & analytics tasks using Tavily Web Search
Identifies repetitive steps in your workflow and sets up Tavily Web Search to handle them automatically