openclaw-skills/tavily-search/SKILL.md
Use when users need real-time web search, fresh facts, source-backed evidence, market/news context, or Tavily-powered retrieval before answering time-sensitive questions.
npx skillsauth add seaworld008/commonly-used-high-value-skills tavily-searchInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
当任务依赖 最新信息(新闻、市场数据、动态变化文档)时,必须优先使用本技能。Tavily Search 专为 AI Agent 设计,相比传统的 Google Search,它能直接返回结构化良好的摘要和高度相关的网页正文片段,显著减少了 Agent 解析无关 HTML 的 Token 开销。
npx clawhub@latest install tavily-search
Sources 引用,增强可信度。tavily_search 接口。search_depth: "advanced" 以获取更深层的抓取。max_results: 5(对于简单概览)或 max_results: 10(对于深度研报)。title, url 和 content 字段。include_raw_content: false 排除不必要的网页源代码,仅保留清洗后的正文。answer 功能获取搜索引擎自动生成的初步简答。web_fetch 针对性抓取特定 URL。include_images: true。url。[^1], [^2] 格式进行上标。References 列表。### 搜索目标 (Search Target)
[描述:查找 NVIDIA 2025 Q4 财报的核心指标及分析师预期]
### 搜索策略 (Search Strategy)
1. **宏观搜索**: 使用词组 `NVIDIA 2025 Q4 earnings report official PDF`.
2. **分析师视角**: 使用词组 `NVIDIA earnings analysis 2025 Goldman Sachs Morgan Stanley`.
3. **市场反应**: 使用词组 `NVDA stock price reaction Q4 2025 post-market`.
### 约束条件 (Constraints)
- 只采集发布日期在 2025 年 1 月之后的信息。
- 优先选择 `.gov`, `.edu` 或知名财经媒体(如 Bloomberg, Reuters)。
### 预期产出 (Desired Output)
- 汇总表:营收、净利、毛利率、同比增长。
- 分析师观点对比。
- 原文来源链接。
// 示例:同时在中文和英文互联网搜索同一技术趋势
mcp_call({
name: 'tavily_search',
arguments: {
query: "DeepSeek-V3 architecture analysis",
search_depth: "advanced",
include_answer: true
}
});
cron 任务,每天早上 8 点调用 tavily-search 抓取特定关键词的新闻,利用 summarize 技能生成摘要,并通过 notion 或 slack 发送给用户。include_domains 或 exclude_domains 过滤掉不靠谱的博客或广告网站。MEMORY.md,避免重复搜索同一个固定事实。development
Enumerating failure modes via pre-mortem analysis. Systematically identifies failure scenarios for plans, designs, and features, scoring them with RPN/AP. Does not write code.
testing
Orchestrating specialist AI agent teams as a meta-coordinator. Decomposes requests into minimum viable chains, spawns each as an independent session in AUTORUN modes, and drives to final output. Use when a task spans multiple specialist domains, requires parallel agent execution, or needs hub-and-spoke routing across the skill ecosystem.
development
Converting document formats (Markdown/Word/Excel/PDF/HTML). Converts specs from Scribe and reports from Harvest into distributable formats; generates reusable conversion scripts. Use when converting documents, building accessibility-compliant PDFs, or creating Pandoc/LibreOffice pipelines.
testing
Curating cross-agent knowledge and guarding institutional memory. Extracts patterns from agent journals into METAPATTERNS.md, detects knowledge decay, propagates best practices, prevents organizational forgetting. Use when consolidating cross-agent insights, curating memory, or auditing knowledge decay.