A powerful search skill that integrates Perplexity Sonar models to provide citation-backed web answers and long-form research reports within your agent workflow.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install aisa-perplexity-search-skill
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 aisa-perplexity-search-skill 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 Perplexity Search?
The Perplexity Search skill empowers your AI agent with real-time web access and advanced synthesis capabilities. By utilizing the Openclaw Skills framework, this tool connects to the AIsa API to provide answers that are not only accurate but also fully cited with live web sources. It is designed for tasks requiring high factual precision, analytical reasoning, and current information.
Whether you need a quick summary of recent events or an exhaustive research report, this skill offers a variety of specialized models to handle different levels of complexity. It bridges the gap between static model training data and the rapidly evolving information available on the internet, making it an essential addition for developers building research-oriented agents.
Perplexity Search Use Cases
- Conducting exhaustive industry research to create deep reports on emerging technologies.
- Verifying facts and gathering citations for technical documentation or articles.
- Comparing products or services using real-time web data for competitive analysis.
- Performing multi-step reasoning tasks that require external verification and analytical depth.
How Perplexity Search Works
- The user or agent triggers a search request specifying the desired query and research depth.
- The skill selects the most appropriate Perplexity model, such as sonar for speed or sonar-deep-research for intensity.
- A request is dispatched via the Python client or curl to the AIsa API endpoint with the required authentication.
- The Perplexity engine crawls the web, synthesizes the findings, and generates a response complete with inline citations.
- The structured Markdown output is returned to the calling agent for immediate use or display.
Perplexity Search Setup
To get started with this skill on Openclaw Skills, ensure you have the necessary dependencies and credentials.
First, set your AIsa API key as an environment variable:
export AISA_API_KEY="your_aisa_api_key_here"
Ensure python3 and curl are installed on your system. You can then use the bundled Python client to execute queries:
python3 {baseDir}/scripts/perplexity_search_client.py sonar-pro --query "Explain the benefits of Openclaw Skills for AI developers"
Perplexity Search Data Schema & Taxonomy
The Perplexity Search skill interacts with a specific set of parameters and returns structured Markdown data. The organization of this data is outlined below:
| Attribute | Type | Description |
|---|---|---|
| Model | String | Selection of sonar, sonar-pro, sonar-reasoning-pro, or sonar-deep-research |
| Query | String | The specific search or research prompt provided by the user |
| System Message | String | Optional formatting instructions for the model output |
| Output | Markdown | Citation-backed text including a list of references and source links |
Perplexity Search Advanced Features
- Access to the Sonar Deep Research model for high-latency, exhaustive analytical reporting.
- Specialized reasoning-pro model for complex, multi-step logical queries that require web verification.
- Customizable system messages to force specific output formats like JSON or executive summaries.
- Built-in retry logic and extended timeout handling for intensive research tasks through the bundled client script.
SKILL.md
Loading
Related Openclaw Skills
HF Papers (Hugging Face Research Explorer)
An AI agent skill to browse trending papers, search research topics, and retrieve detailed metadata from the Hugging Face Papers platform.
Break-even Calculator for Ecommerce
A professional-grade unit economics tool designed to calculate the real no-loss line for ecommerce products by modeling complex variable costs.
Bilibili Hot Trend
A specialized automation tool for analyzing Bilibili video performance, creator metrics, and public ranking trends.
Camera API Skill
A comprehensive AI skill for managing cloud-based cameras, enabling real-time monitoring, screenshot capture, and automated event summarization.
Ubuntu Inspector
A comprehensive system inspection and health monitoring tool designed for Ubuntu servers.
OpenGraph.io
An essential integration for extracting Open Graph metadata, titles, and images from any URL to enhance link previews.
METADATA
Featured*
