exa skill

safe

This skill helps you perform neural web searches and extract code context using the Exa API to locate docs, code snippets, and papers.

npx playbooks add skill openclaw/skills --skill exa

Review the files below or copy the command above to add this skill to your agents.

Files (5)

SKILL.md

957 B

---
name: exa
description: Neural web search and code context via Exa AI API. Requires EXA_API_KEY. Use for finding documentation, code examples, research papers, or company info.
metadata: {"clawdbot":{"emoji":"🧠","requires":{"env":["EXA_API_KEY"]}}}
---

# Exa - Neural Web Search

Direct API access to Exa's neural search engine.

## Setup

**1. Get your API Key:**
Get a key from [Exa Dashboard](https://dashboard.exa.ai/api-keys).

**2. Set it in your environment:**
```bash
export EXA_API_KEY="your-key-here"
```

## Usage

### Web Search
```bash
bash scripts/search.sh "query" [num_results] [type]
```
*   `type`: auto (default), neural, fast, deep
*   `category`: company, research-paper, news, github, tweet, personal-site, pdf

### Code Context
Finds relevant code snippets and documentation.
```bash
bash scripts/code.sh "query" [num_results]
```

### Get Content
Extract full text from URLs.
```bash
bash scripts/content.sh "url1" "url2"
```

Overview

This skill provides neural web search and code-context retrieval via the Exa AI API. It lets you find documentation, code examples, research papers, company info, and extract full page content. The skill requires an EXA_API_KEY set in your environment to authenticate requests.

How this skill works

The skill calls Exa's neural search endpoints to return ranked results across categories like research papers, GitHub, news, and personal sites. It can run targeted searches (neural, fast, deep) and retrieve code-relevant snippets or full-page text from URLs for downstream analysis. Scripts provide command-line entry points for search, code context, and content extraction.

When to use it

You need high-quality, relevance-ranked results across heterogeneous web sources.
You want concise code examples and surrounding documentation for a programming question.
You need to locate research papers, company profiles, or GitHub repositories quickly.
You want to extract full text from one or more URLs for indexing or summarization.
You are building tools that require neural semantic search rather than keyword-only search.

Best practices

Set EXA_API_KEY in your environment before running any scripts to avoid auth errors.
Start with 'type=auto' for general searches, switch to 'neural' or 'deep' for stronger semantic matches.
Limit result counts when prototyping to speed up responses and reduce API usage.
Filter by category (e.g., github, research-paper, pdf) to narrow results and improve relevance.
Extract full content only when you need raw text for processing to save bandwidth and time.

Example use cases

Find implementation examples for a library function across GitHub and personal blogs.
Locate the latest research papers on a topic and extract PDFs or full-text snippets.
Pull company background and news mentions for a due-diligence summary.
Gather code snippets and contextual docs to feed into an automated code-assistant pipeline.
Bulk-extract page content to build an offline archive or searchable index of resources.

FAQ

How do I authenticate?

Set your API key in the EXA_API_KEY environment variable (export EXA_API_KEY="your-key").

What search types are available?

The skill supports type values auto, neural, fast, and deep to control retrieval behavior.

Can I restrict searches to a source type?

Yes. Use category filters like company, research-paper, github, news, tweet, personal-site, or pdf.