investment-research-briefs

作者: tavily-ai

生成简洁的投资研究简报、公司备忘录、行业快照、投资组合监控更新、财报/新闻摘要、风险与催化剂简报……

npx skills add https://github.com/tavily-ai/use-case-skills --skill investment-research-briefs

Investment Research Briefs

Workflow

Use search and extract for quick investor briefs and source checks, research for full investment memos, and map or crawl only for large known source collections. Keep this skill focused on research design, source selection, verification, and synthesis; execution mechanics should come from companion endpoint skills.

Treat the guidance below as base guidance; adapt it to the user's request when appropriate.

  • Define the target, investor lens, geography, timeframe, asset type, and desired depth.
  • For quick briefs, split into short subqueries under 400 characters: business overview, market position, financial/operating signals, recent developments, competitors, risks, catalysts, and valuation/comps if requested.
  • Prefer filings, investor relations pages, earnings materials, company pages, regulators, exchanges, reputable financial media, and market sources.
  • For deeper memos, provide a clear research goal, known context, constraints, target market, and desired output shape.
  • Verify specific metrics, quotes, management claims, filings, and high-impact assertions against original or authoritative sources.
  • Separate sourced facts, inferred interpretation, and open diligence questions.

Research Budget

  • Start with a small focused search set covering business context, recent developments, financial/operating signals, and risks/catalysts.
  • Extract only the strongest sources before drafting a concise brief.
  • Add more searches only for named gaps, such as missing filing evidence, missing earnings context, or missing competitor context.
  • Do not use map unless a known investor-relations, filing, transcript, or presentation library has buried pages.
  • Do not use research unless the user asks for a full memo, sector landscape, or multi-source thesis.

Capability Guidance

  • Use search for fast discovery across companies, sectors, filings, news, competitors, and market context.
  • Use extract to verify metrics, quotes, filings, management claims, transcripts, and other high-impact assertions.
  • Use research for full investment memos, sector landscapes, or multi-source thesis work.
  • Use map or crawl only for large investor-relations, filing, transcript, or presentation libraries.

Query And Source Guidance

  • Use company legal name, ticker, product names, segment names, geography, competitor names, and reporting period in queries.
  • For public companies, prioritize original filings, earnings calls, investor presentations, regulator pages, and exchange notices.
  • For private companies, triangulate from company pages, funding announcements, customer stories, hiring, product documentation, and credible press.
  • For recent developments, make the timeframe explicit in the query.
  • For competitor or sector work, build separate subqueries for category, demand drivers, pricing, regulation, customer adoption, and risks.
  • Include known assumptions or prior context in broad research requests to avoid rediscovering what is already known.
  • Report failed or unavailable sources when they affect filings, transcripts, IR pages, market data, or cited claims.

Output Template

Use this markdown structure and adapt sections to the user's mandate:

# Investment Brief: <company/sector/theme>

## Executive Summary
<concise bullets with the main investor-relevant takeaways>

## Business And Market Context
- Business model:
- Customers/segments:
- Market position:
- Competitors:

## Recent Developments
- <dated development with source>

## Signals
- Financial or operating signals:
- Product, customer, hiring, or partnership signals:
- Regulatory or macro signals:

## Risks
- <risk>: <evidence and uncertainty>

## Catalysts And Watch Items
- <catalyst/watch item>: <why it matters>

## Open Diligence Questions
- <question>

## Sources
- [Source title](URL) - <what it supports>

Do not provide personalized financial advice. State when source coverage is thin, stale, or incomplete.

来自 tavily-ai 的更多技能

research
tavily-ai
针对任意主题进行综合研究,自动收集来源、分析并生成引用。通过多源网络研究并附带明确引用,适用于对比分析、时事追踪、市场调研及详细报告。提供三种模型选项:mini用于针对性单主题研究(约30秒),pro用于全面多角度分析(约60-120秒),auto通过API自动检测复杂度。通过Tavily MCP服务器的OAuth进行身份验证,支持基于浏览器的自动登录...
official
search
tavily-ai
基于LLM优化的网页搜索,具备相关性评分与灵活筛选功能。支持四种搜索深度模式(极速、快速、基础、高级),可配置延迟与相关性权衡。包含域名过滤、时间范围限制、日期区间、国家加权及原始内容提取。返回结果含标题、URL、内容摘要及相关性评分;可选图片结果与网站图标。通过Tavily MCP服务器或API密钥配置实现自动OAuth认证;...
official
tavily-best-practices
tavily-ai
面向LLM的网页搜索API,支持实时数据访问、内容提取、站点爬取及AI驱动研究。五大核心方法:search()获取网页结果,extract()提取URL内容,crawl()进行全站提取,map()发现URL,research()实现端到端AI综合。提供Python和JavaScript SDK,支持异步客户端并行查询及可配置搜索深度(极速/快速/基础/高级)。crawl方法接受语义指令以聚焦提取内容...
official
tavily-cli
tavily-ai
通过Tavily CLI实现网页搜索、内容提取、站点爬取及深度研究。五种命令模式涵盖搜索、提取、URL发现、批量爬取及带引用的多源研究。所有命令支持JSON输出及文件保存,适用于结构化、智能体工作流。升级模式根据需求引导您从简单搜索逐步过渡到提取、映射、爬取乃至全面研究。需安装tavily-cli并通过tvly login进行API密钥认证。
official
tavily-crawl
tavily-ai
多页网站爬虫,具备语义过滤和Markdown导出功能。可控制深度和广度爬取整个网站部分;通过路径正则表达式、域名或自然语言指令过滤结果,聚焦所需内容。使用--output-dir参数将每个页面保存为本地Markdown文件,或返回结构化JSON供智能体处理。采用语义指令与分块提取,防止向LLM输入结果时上下文膨胀;支持全页提取用于离线文档下载。支持...
official
tavily-dynamic-search
tavily-ai
搜索网络、过滤结果并提取内容,使原始搜索数据永远不会进入您的上下文窗口。只有您精心整理的print()输出会返回。
official
tavily-extract
tavily-ai
从最多20个URL中提取干净的Markdown或文本,支持JavaScript渲染和查询聚焦分块。可处理JavaScript渲染页面,提取深度可配置(简单页面使用基础模式,动态SPA和表格使用高级模式)。支持查询聚焦提取,仅返回相关的内容块而非完整页面。默认返回经LLM优化的Markdown格式,也可选择纯文本格式和结构化JSON输出。单次调用最多处理20个URL;...
official
tavily-map
tavily-ai
快速发现网站上的URL而不提取内容,适用于在大型网站上查找特定页面。返回域内所有URL的结构化列表,支持可配置的深度和广度、正则表达式路径过滤,以及用于语义过滤的自然语言指令。支持深度控制(1-5层)、每页广度限制、外部链接的包含/排除,以及通过正则表达式模式进行域过滤。设计为工作流程中的第一步:通过映射找到正确页面,然后使用提取或...
official