home / skills / phy041 / claude-agent-skills / exa-web-search-free
This skill helps you perform free web, code, and company research using Exa MCP, delivering quick insights without API keys.
npx playbooks add skill phy041/claude-agent-skills --skill exa-web-search-freeReview the files below or copy the command above to add this skill to your agents.
---
name: exa-web-search-free
description: Free AI-powered search via Exa MCP. Web search for news/info, code search for docs/examples from GitHub/StackOverflow, company research for business intel. No API key needed.
metadata: {"openclaw":{"emoji":"🔍","requires":{"bins":["mcporter"]}}}
inputs:
- name: query
type: string
required: true
description: The search query string
- name: search_type
type: string
required: false
description: Type of search — "web" (default), "code" (GitHub/StackOverflow), or "company" (business intel)
outputs:
- name: results
type: list
description: List of search results, each with {title, url, snippet}
---
# Exa Web Search (Free)
Neural search for web, code, and company research. No API key required.
## Setup
Verify mcporter is configured:
```bash
mcporter list exa
```
If not listed:
```bash
mcporter config add exa https://mcp.exa.ai/mcp
```
## Core Tools
### web_search_exa
Search web for current info, news, or facts.
```bash
mcporter call 'exa.web_search_exa(query: "latest AI news 2026", numResults: 5)'
```
**Parameters:**
- `query` - Search query
- `numResults` (optional, default: 8)
- `type` (optional) - `"auto"`, `"fast"`, or `"deep"`
### get_code_context_exa
Find code examples and docs from GitHub, Stack Overflow.
```bash
mcporter call 'exa.get_code_context_exa(query: "React hooks examples", tokensNum: 3000)'
```
**Parameters:**
- `query` - Code/API search query
- `tokensNum` (optional, default: 5000) - Range: 1000-50000
### company_research_exa
Research companies for business info and news.
```bash
mcporter call 'exa.company_research_exa(companyName: "Anthropic", numResults: 3)'
```
**Parameters:**
- `companyName` - Company name
- `numResults` (optional, default: 5)
## Advanced Tools (Optional)
Six additional tools available by updating config URL:
- `web_search_advanced_exa` - Domain/date filters
- `deep_search_exa` - Query expansion
- `crawling_exa` - Full page extraction
- `people_search_exa` - Professional profiles
- `deep_researcher_start/check` - AI research agent
**Enable all tools:**
```bash
mcporter config add exa-full "https://mcp.exa.ai/mcp?tools=web_search_exa,web_search_advanced_exa,get_code_context_exa,deep_search_exa,crawling_exa,company_research_exa,people_search_exa,deep_researcher_start,deep_researcher_check"
# Then use:
mcporter call 'exa-full.deep_search_exa(query: "AI safety research")'
```
## Tips
- Web: Use `type: "fast"` for quick lookup, `"deep"` for thorough research
- Code: Lower `tokensNum` (1000-2000) for focused, higher (5000+) for comprehensive
## Resources
- [GitHub](https://github.com/exa-labs/exa-mcp-server)
- [npm](https://www.npmjs.com/package/exa-mcp-server)
- [Docs](https://exa.ai/docs)
This skill provides free AI-powered web, code, and company search through Exa MCP with no API key required. It surfaces current news, documentation and code examples from GitHub/Stack Overflow, and business intelligence on companies. The skill is tailored for founders, indie hackers, and growth engineers who need fast, practical research workflows. Advanced optional tools enable deeper crawling, filtering, and professional profile lookups.
The skill calls Exa MCP endpoints to run neural search across web pages, code repositories, Q&A sites, and company data. Use web_search_exa for news and facts, get_code_context_exa for code samples and docs, and company_research_exa for business intelligence. Optional advanced tools add domain/date filters, query expansion, full-page crawling, and people search for more targeted or exhaustive investigations.
Do I need an API key to use this skill?
No. The Exa MCP-based skill is designed to work without an API key when configured correctly via mcporter.
How do I control search depth and speed?
Use the type parameter ("fast" or "deep") for web searches and adjust tokensNum for code context to trade off speed versus thoroughness.