Exa Search

1---2name: exa-search3description: Neural search via Exa MCP for web, code, and company research. Use when the user needs web search, code examples, company intel, people lookup, or AI-powered deep research with Exa's neural search engine.4origin: ECC5---6 7# Exa Search8 9Neural search for web content, code, companies, and people via the Exa MCP server.10 11## When to Activate12 13- User needs current web information or news14- Searching for code examples, API docs, or technical references15- Researching companies, competitors, or market players16- Finding professional profiles or people in a domain17- Running background research for any development task18- User says "search for", "look up", "find", or "what's the latest on"19 20## MCP Requirement21 22Exa MCP server must be configured. Add to `~/.claude.json`:23 24```json25"exa-web-search": {26  "command": "npx",27  "args": ["-y", "exa-mcp-server"],28  "env": { "EXA_API_KEY": "YOUR_EXA_API_KEY_HERE" }29}30```31 32Get an API key at [exa.ai](https://exa.ai).33This repo's current Exa setup documents the tool surface exposed here: `web_search_exa` and `get_code_context_exa`.34If your Exa server exposes additional tools, verify their exact names before depending on them in docs or prompts.35 36## Core Tools37 38### web_search_exa39General web search for current information, news, or facts.40 41```42web_search_exa(query: "latest AI developments 2026", numResults: 5)43```44 45**Parameters:**46 47| Param | Type | Default | Notes |48|-------|------|---------|-------|49| `query` | string | required | Search query |50| `numResults` | number | 8 | Number of results |51| `type` | string | `auto` | Search mode |52| `livecrawl` | string | `fallback` | Prefer live crawling when needed |53| `category` | string | none | Optional focus such as `company` or `research paper` |54 55### get_code_context_exa56Find code examples and documentation from GitHub, Stack Overflow, and docs sites.57 58```59get_code_context_exa(query: "Python asyncio patterns", tokensNum: 3000)60```61 62**Parameters:**63 64| Param | Type | Default | Notes |65|-------|------|---------|-------|66| `query` | string | required | Code or API search query |67| `tokensNum` | number | 5000 | Content tokens (1000-50000) |68 69## Usage Patterns70 71### Quick Lookup72```73web_search_exa(query: "Node.js 22 new features", numResults: 3)74```75 76### Code Research77```78get_code_context_exa(query: "Rust error handling patterns Result type", tokensNum: 3000)79```80 81### Company or People Research82```83web_search_exa(query: "Vercel funding valuation 2026", numResults: 3, category: "company")84web_search_exa(query: "site:linkedin.com/in AI safety researchers Anthropic", numResults: 5)85```86 87### Technical Deep Dive88```89web_search_exa(query: "WebAssembly component model status and adoption", numResults: 5)90get_code_context_exa(query: "WebAssembly component model examples", tokensNum: 4000)91```92 93## Tips94 95- Use `web_search_exa` for current information, company lookups, and broad discovery96- Use search operators like `site:`, quoted phrases, and `intitle:` to narrow results97- Lower `tokensNum` (1000-2000) for focused code snippets, higher (5000+) for comprehensive context98- Use `get_code_context_exa` when you need API usage or code examples rather than general web pages99 100## Related Skills101 102- `deep-research` — Full research workflow using firecrawl + exa together103- `market-research` — Business-oriented research with decision frameworks