Perplexity Deep Search for Openclaw

An AI-powered web search tool providing three levels of depth—quick search, logical reasoning, and exhaustive research—directly through the Perplexity API.

ericsantos
v1.1.0
Feb 19, 2026
14
5.2k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install perplexity-deep-search

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install perplexity-deep-search 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 Deep Search?

The Perplexity Deep Search skill brings the power of real-time web intelligence to your AI coding agent. By leveraging specialized models like sonar-pro and sonar-reasoning-pro, this Openclaw Skills entry enables agents to ground their responses in up-to-date facts and verified citations.

This integration allows developers and researchers to bypass manual browsing by automating the retrieval of high-signal information. Whether you need a quick summary of current events or a multi-step logical analysis of complex technical architectures, this skill provides a structured interface for high-quality information retrieval directly within your development environment.

Perplexity Deep Search Use Cases

  • Gathering the latest API documentation or software release notes.
  • Comparing competing technologies or frameworks for enterprise decision-making.
  • Generating exhaustive market analysis and literature reviews with the research mode.
  • Restricting searches to specific academic or technical domains for higher accuracy.

How Perplexity Deep Search Works

  1. The user executes the search script using the CLI, specifying the query and optional flags for mode, recency, or domain filtering.
  2. The skill validates the PERPLEXITY_API_KEY and confirms the presence of system dependencies like curl and jq.
  3. The request is routed to the selected Perplexity model (Sonar, Sonar Reasoning, or Deep Research) based on the specified depth.
  4. The AI processes the query against real-time web data, synthesizing a grounded answer with supporting citations.
  5. The final response is returned as formatted Markdown or raw JSON for further processing by the agent.

Perplexity Deep Search Setup

To use this skill within the Openclaw Skills ecosystem, ensure you have curl and jq installed.

Set your API key as an environment variable:

export PERPLEXITY_API_KEY="your_api_key_here"

Alternatively, you can store the key in a local configuration file for persistence:

mkdir -p ~/.config/perplexity
echo "your_key_here" > ~/.config/perplexity/api_key
chmod 600 ~/.config/perplexity/api_key

Perplexity Deep Search Data Schema & Taxonomy

The skill organizes its output based on the selected mode and provided flags. It supports structured parameters to narrow the scope of the search.

Option Description Values
--mode Selection of search depth search, reason, research
--recency Temporal filter for results hour, day, week, month
--domains Domain-specific filtering Comma-separated domains
--lang Output language en, pt, es, etc.
--json Data output format Boolean (on/off)

Perplexity Deep Search Advanced Features

  • Triple-tier search modes allowing you to balance query cost with analysis depth.
  • Domain restriction capabilities to ensure the AI only pulls data from trusted repositories like GitHub or official documentation.
  • Deep Research mode for generating comprehensive reports that utilize multiple sub-queries and reasoning steps.
  • Native JSON output support for piping search results into other automation scripts or data analysis tools.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*