菜单
Neural web search via Exa AI. Search people, companies, news, research, code. Supports deep search, domain filters, date ranges.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
基于 SOC 职业分类
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/lev-os/agents --skill exa-plus命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
正在显示 SKILL.md
| name | exa-plus |
| version | 1.0.0 |
| description | Neural web search via Exa AI. Search people, companies, news, research, code. Supports deep search, domain filters, date ranges. |
| metadata | {"clawdbot":{"emoji":"🧠","requires":{"bins":"[Truncated]"}}} |
Powerful AI-powered search with LinkedIn, news, research papers, and more.
Set EXA_API_KEY in your environment (e.g. ~/.env.local):
export EXA_API_KEY="your-exa-api-key"
Fallback: ~/.clawdbot/credentials/exa/config.json with {"apiKey": "..."}
bash scripts/search.sh "query" [options]
Options (as env vars):
NUM=10 - Number of results (max 100)TYPE=auto - Search type: auto, neural, fast, deepCATEGORY= - Category: news, company, people, research paper, github, tweet, pdf, financial reportDOMAINS= - Include domains (comma-separated)EXCLUDE= - Exclude domains (comma-separated)SINCE= - Published after (ISO date)UNTIL= - Published before (ISO date)LOCATION=NL - User location (country code)# Basic search
bash scripts/search.sh "AI agents 2024"
# LinkedIn people search
CATEGORY=people bash scripts/search.sh "software engineer Amsterdam"
# Company search
CATEGORY=company bash scripts/search.sh "fintech startup Netherlands"
# News from specific domain
CATEGORY=news DOMAINS="reuters.com,bbc.com" bash scripts/search.sh "Netherlands"
# Research papers
CATEGORY="research paper" bash scripts/search.sh "transformer architecture"
# Deep search (comprehensive)
TYPE=deep bash scripts/search.sh "climate change solutions"
# Date-filtered news
CATEGORY=news SINCE="2026-01-01" bash scripts/search.sh "tech layoffs"
Extract full text from URLs:
bash scripts/content.sh "url1" "url2"
exa-plus specializes in neural search (people, companies, research). For other needs:
| Tool | Specialty | Use When |
|---|---|---|
| exa-plus (this) | Neural web, GitHub, LinkedIn, papers | People/company search, academic research |
| valyu | Recursive turn-based research | valyu research "query" --turns 5 |
| deep-research | Multi-query Tavily synthesis | deep-research "query" --deep |
| lev-research | Multi-perspective orchestration | lev-research "query" |
| lev-find | Unified local + external | lev find "query" --scope=research |
| brave-search | Quick web search | brave-search "query" |
| tavily-search | AI-optimized snippets | tavily-search "query" |
| grok-research | Real-time X/Twitter | grok-research "query" |
| firecrawl | Web scraping | firecrawl scrape <url> |
| qmd | Local session search | qmd query "query" |
Exa's unique capabilities:
CATEGORY=people)CATEGORY=company)CATEGORY="research paper")CATEGORY=github)Integration pattern:
# 1. Find people/companies with Exa
CATEGORY=people exa search "ML engineer Amsterdam"
# 2. Deep research on findings
valyu research "machine learning hiring trends" --turns 5
See skill://lev-research for comprehensive research workflows.
Routes adaptive multi-agent deliberation with fractal context cycles. Use when using /cdo, think/deep/debug/parliament work, or long runs paired with autoresearch scheduling.
Use when you need an evidence-first provenance report from `cass`, repo docs, `.lev`, `~/.agents/diagrams`, `qmd`, or Grep/Glob tools.
Use the codebase knowledge graph for structural code queries. Triggers on: explore the codebase, understand the architecture, what functions exist, show me the structure, who calls this function, what does X call, trace the call chain, find callers of, show dependencies, impact analysis, dead code, unused functions, high fan-out, refactor candidates, code quality audit, graph query syntax, Cypher query examples, edge types, how to use search_graph.