A high-performance neural search engine integration that allows AI agents to perform deep, category-specific web research and full-text content extraction.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install exa-plus
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install exa-plus using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Exa Plus Neural Search?
Exa Plus is an advanced neural search skill designed to empower AI agents with the ability to crawl and understand the web like a human. Unlike traditional keyword-based search engines, this Openclaw Skills integration leverages Exa AI's embeddings to find content based on meaning and context, making it ideal for complex research tasks that require more than just keyword matching.
By integrating this skill, developers can enable their agents to filter results by specific categories such as LinkedIn profiles, financial reports, or GitHub repositories. It provides the necessary infrastructure to not only discover relevant URLs but also extract their full text content, enabling a seamless workflow from discovery to data processing within any agentic framework.
Exa Plus Neural Search Use Cases
- Conducting deep market research on specific industries or emerging technologies.
- Finding talent or subject matter experts by searching LinkedIn profiles directly through an AI agent.
- Monitoring news and social media trends using precise date-range filters.
- Gathering academic sources and research papers for technical documentation and whitepapers.
- Automating competitive analysis by extracting data from specific high-authority domains.
How Exa Plus Neural Search Works
- The skill receives a search query along with optional environment variables like category, date range, or domain filters.
- It sends a secure request to the Exa AI API using neural search algorithms to identify the most relevant pages based on semantic similarity.
- The results are returned to the agent in a structured format, including titles, URLs, and metadata for evaluation.
- If full text is required for analysis, the content extraction script fetches the readable text from the identified URLs to be processed by the LLM.
Exa Plus Neural Search Setup
To get started with this skill in your Openclaw Skills environment, ensure you have curl and jq installed on your system. Then, configure your credentials as follows:
- Create the configuration directory:
mkdir -p ~/.clawdbot/credentials/exa/
- Create the config.json file and add your Exa API key:
echo '{"apiKey": "your-exa-api-key"}' > ~/.clawdbot/credentials/exa/config.json
- You can now run searches using the provided shell scripts in the scripts/ directory.
Exa Plus Neural Search Data Schema & Taxonomy
The skill manages search parameters and output through a structured schema designed for Openclaw Skills interoperability. Data is typically handled as environment variables during script execution.
| Parameter | Description | Possible Values |
|---|---|---|
| NUM | Number of results to return | 1 to 100 |
| TYPE | Search algorithm depth | auto, neural, fast, deep |
| CATEGORY | Semantic grouping | news, company, people, research paper, github, tweet, pdf, financial report |
| DOMAINS | Inclusion whitelist | Comma-separated domains (e.g., reuters.com) |
| SINCE/UNTIL | Date filtering | ISO 8601 formatted dates (e.g., 2024-01-01) |
Exa Plus Neural Search Advanced Features
- Deep Search Mode: Employs more intensive neural processing to find high-quality results for complex or niche queries.
- Domain-Specific Filtering: Whitelist or blacklist specific website clusters to ensure data quality and source reliability.
- Semantic Categorization: Automatically restricts the search scope to specific data types like GitHub repositories or LinkedIn profiles without complex query syntax.
- Batch Content Extraction: Built-in capability to pull raw, readable text from multiple URLs simultaneously for immediate processing by the AI agent.
SKILL.md
Loading
Related Openclaw Skills
Ikuzo Travel & Map Management
A comprehensive map and travel planning skill for managing locations, spots, and daily itineraries.
YouTube Thumbnail Grabber
A versatile tool for downloading YouTube video thumbnails in multiple resolutions and formats using yt-dlp.
YouTube Video Summarizer
A powerful utility to programmatically extract, clean, and prepare YouTube transcripts for AI-driven summarization and content analysis.
Yahoo Finance
A powerful financial data integration for retrieving real-time stock prices, historical market trends, and detailed company analytics.
Ralph Loop
A framework to generate autonomous bash scripts that run AI coding agents in continuous planning and building loops.
Sonarr
An automation skill for searching and managing TV show libraries within Sonarr via AI agent commands.
Featured*
