exa-web-search
Use the Exa MCP server for internet search and full-page retrieval. Use whenever the user asks to search, browse, look up, verify, research, find so…
它会碰到什么
扫了多少2 个文本文件,2 KB
它会碰到什么不碰外部(只输出文字)
命中总数0 处
命中统计严重 0 · 高 0 · 中 0 · 低 0
这一栏是扫描器报的事实,不是结论。命中多不等于有毒(安全工具、规则库、示例脚本本来就会包含危险写法),命中少也不等于干净。它和你手上的凭据、文件、网络有什么关系,需要你自己看。
技能内容
Exa Web Search
Use Exa as the first web-search tool for matching requests.
Workflow
- Call
mcp__exa__web_search_exabefore drafting an answer that depends on online information. - Write a natural-language query describing the ideal source. Include dates, domains, or source types when they materially narrow the task.
- Request enough results to compare sources; use 5-10 for broad research and fewer for a targeted lookup.
- Call
mcp__exa__web_fetch_exaon the best URLs when search excerpts do not contain enough context, when exact attribution matters, or when the user asks to read a specific page. - For time-sensitive or consequential claims, compare multiple independent or primary sources. Prefer official documentation, original papers, first-party announcements, and source repositories.
- Cite the final source URLs close to the claims they support. Distinguish source statements from inference.
Boundaries
- Honor explicit user constraints about tools, domains, dates, languages, and source types.
- Use
rgor repository tools for local files instead of Exa. - Prefer a dedicated connected system when it is the authoritative source, such as Zotero for the user's library, Figma for a design file, or OpenAI Developer Docs for OpenAI product documentation.
- If Exa is unavailable or returns inadequate results, state that briefly and use another available web-search tool so the task still completes.
想直接用这个技能?
本站把开放许可(MIT / Apache 等)的技能按仓库打包整理到网盘,点一下转存到你自己的网盘,不用一个个从 GitHub 拉。许可未声明的技能只给原始仓库链接,不打包。
它属于哪个仓库
星标★ 1,027
本站分层T1
该仓技能数1910
原文件路径
plugins/zlsbksdxl/codex-exa/skills/exa-web-search/SKILL.md