perplexica-search
AI-powered search using your local Perplexica instance. Combines web search with LLM reasoning for accurate answers with cited sources. Use when user asks to "search with Perplexica", "ask Perplexica", "deep search", "research with sources", or wants AI search with citations.
Why use this skill?
Integrate Perplexica with OpenClaw for local, privacy-focused AI research. Get cited search results, deep insights, and smart summaries directly in your terminal.
Install via CLI (Recommended)
clawhub install openclaw/skills/skills/eplt/perplexica-search-localWhat This Skill Does
The perplexica-search skill integrates your local Perplexica instance directly into OpenClaw, transforming your environment into an AI-powered research hub. By leveraging your own local deployment, this skill performs web searches and uses LLM reasoning to synthesize accurate answers complete with verifiable citations. Unlike standard web scrapers, it handles complex queries by evaluating context and providing evidence-backed responses based on your configured models.
Installation
To install this skill, run the following command in your terminal:
clawhub install openclaw/skills/skills/eplt/perplexica-search-local
Ensure you have a Perplexica instance running locally (typically at http://localhost:3000) and that your chat and embedding models are properly configured within your Perplexica dashboard before executing the first search.
Use Cases
This skill is designed for scenarios requiring high-fidelity information retrieval. It is ideal for:
- Technical Research: Digging into documentation, GitHub discussions, or StackOverflow threads without leaving your CLI.
- Academic Investigation: Finding and summarizing peer-reviewed academic papers.
- Privacy-First Intelligence: Conducting deep web research without telemetry, keeping your queries local to your infrastructure.
- Content Synthesis: Generating reports or summaries that require multiple sources to validate claims.
Example Prompts
- "Search with Perplexica: What are the current industry standard approaches for RAG architecture?"
- "Research with sources: Compare the pros and cons of using Vector DBs vs Graph DBs for LLM context injection."
- "Deep search: Find recent updates on the implementation of memory safety in the Rust programming language."
Tips & Limitations
- Model Selection: If the tool returns errors, check that your
chat-modelandembedding-modelkeys match those exactly configured in your Perplexica instance. - Performance: 'Quality' mode produces the best output but is significantly slower; use 'balanced' or 'speed' for quick lookups.
- System Instructions: Use the
-iflag to guide the AI's persona, such as "act as a senior software architect," to refine the depth and tone of the responses. - Network Dependence: This skill requires an active internet connection on the machine where Perplexica is hosted; it will not function offline.
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-eplt-perplexica-search-local": {
"enabled": true,
"auto_update": true
}
}
}Tags(AI)
Flags: network-access, external-api