Agent Skills
› composio-community/awesome-agent-clis
› Tavily CLI
Tavily CLI
GitHub提供终端内的自然语言搜索、URL内容提取、目标导向爬取及多源深度研究功能，输出结构化JSON供Agent直接解析，专为AI工作流设计。
Trigger Scenarios
需要通过网络搜索获取最新信息或特定数据
需要从指定URL提取结构化网页内容
需要进行跨网站的多源深度调研
需要按特定目标爬取网站信息
Install
npx skills add composio-community/awesome-agent-clis --skill Tavily CLI -g -y
SKILL.md
Frontmatter
{
"name": "Tavily CLI",
"description": "Web search, page extraction, site crawling, and multi-source research from the terminal. Use when an agent needs to search the web, extract content from URLs, crawl sites with an objective, or run deep research across sources."
}
Tavily CLI
Native web access for agents from the terminal. Search, extract, crawl, and research with structured output designed for agent workflows.
- Website: https://tavily.com
- Docs: https://docs.tavily.com
Installation
curl -fsSL https://cli.tavily.com/install.sh | bash
tvly login
Key Commands
Search
Natural language web search — results are ranked, filtered, and compressed for LLM use (no raw HTML).
tvly search "latest developments in AI agents"
tvly search "Series B AI infrastructure companies founded after 2023" --json
Extract
Pull structured content from a specific URL.
tvly extract "https://docs.example.com/api" --json
Crawl
Crawl a site with a specific objective.
tvly crawl "https://docs.tavily.com/welcome" --objective "find documentation on n8n integration" --json
Research
Deep multi-source research on a topic.
tvly research "competitive landscape of AI code assistants" --json
Output Modes
- Default: human-readable
--json: structured JSON output for agent parsing
Agent-Friendly Features
- All commands support
--jsonfor structured output - Results pre-processed for LLM consumption (no raw HTML cleanup needed)
- Designed for iterative search loops (search → inspect → refine → search again)
- API key auth via
tvly login
Version History
- 9f765d2 Current 2026-07-24 16:04
