perplexity
Deep search via Perplexity API. Three modes: search (quick facts), reason (complex analysis), research (in-depth reports). Returns AI-grounded answers with citations.
Why use this skill?
Power your OpenClaw agent with Perplexity-backed search. Get grounded, cited, and intelligent answers ranging from quick facts to deep market research reports.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/ericsantos/perplexity-deep-searchWhat This Skill Does
The Perplexity Deep Search skill integrates advanced, AI-powered web searching directly into your OpenClaw workflow. Unlike standard search tools that provide simple links, this agent utilizes the Perplexity API to deliver grounded, synthesized answers supported by real-time web data and source citations. It features three specialized execution modes tailored to specific cognitive depths: 'search' for rapid information retrieval, 'reason' for complex logical deduction and comparisons, and 'research' for extensive, long-form report generation.
Installation
To integrate this capability into your environment, use the OpenClaw command-line interface:
clawhub install openclaw/skills/skills/ericsantos/perplexity-deep-search
Post-installation, ensure you have a valid API key. You can provide this via the PERPLEXITY_API_KEY environment variable or by creating a configuration file at ~/.config/perplexity/api_key with appropriate permissions (600). The agent is designed to automatically detect your key during execution.
Use Cases
- Software Engineering: Querying the latest documentation for rapidly evolving frameworks or debugging errors by searching specific technical domains like arxiv.org.
- Content Strategy: Compiling comprehensive market analysis reports, competitive benchmarking, or literature reviews without manually navigating hundreds of search results.
- Daily Operations: Efficiently summarizing news events, verifying facts with source-backed citations, or solving multi-step logical problems that require external context.
Example Prompts
- "Perform a deep research report on the current state of edge computing in industrial IoT for 2025 using the research mode."
- "Compare the performance trade-offs between Tailwind CSS and CSS Modules for a high-traffic e-commerce platform using the reasoning mode."
- "Summarize the latest developments in large language model evaluation benchmarks from the last week and provide links to the source papers."
Tips & Limitations
To optimize costs, default to the 'search' mode ($0.01/query) for routine tasks. Reserve the 'research' mode ($0.40/query) for deep-dive requirements, as it executes multiple, iterative search passes. Be mindful of domain filtering (e.g., --domains "arxiv.org,nature.com") to ensure high-signal results when performing technical or scientific inquiries. Because this tool relies on external web browsing, it requires an active internet connection to function.
Metadata
Not sure this is the right skill?
Describe what you want to build — we'll match you to the best skill from 16,000+ options.
Find the right skillPaste this into your clawhub.json to enable this plugin.
{
"plugins": {
"official-ericsantos-perplexity-deep-search": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, external-api
Related Skills
serpapi
Search Google via SerpAPI (Google Search, Google News, Google Local). Use when you need to search the web, find news articles, or look up local businesses. Supports country/language targeting for region-specific results.
relay-to-agent
Relay messages to AI agents on any OpenAI-compatible API. Supports multi-turn conversations with session management. List agents, send messages, reset sessions.
jina-reader
Web content extraction via Jina AI Reader API. Three modes: read (URL to markdown), search (web search + full content), ground (fact-checking). Extracts clean content without exposing server IP.
show-my-ip
Show the current public IP address of the server. Use when asked about IP, public IP, or network identity.
messaging
NexusMessaging Protocol client — ephemeral agent-to-agent sessions. Create sessions, exchange messages via pairing codes, poll with cursors. Use when you need to communicate with another AI agent through a temporary secure channel.