MCP-perplexica

*"Stop asking Google. Start discovering."*

This little server bridges the gap between Perplexica and any MCP-compatible AI assistant. Think of it as your research companion that never sleeps, never judges your weird search queries, and actually understands what you're trying to find.

Perplexica is an open-source AI-powered search engine that actually understands your questions. Unlike traditional search that throws keywords at you, Perplexica uses AI to dig into the web and find what you actually need — with sources, no less.

• Source: github.com/ItzCrazyKns/Vane (yes, it got renamed from Perplexica to Vane, don't ask)
• Privacy-first: Uses SearXNG to avoid tracking
• Multiple search modes: Web, academic papers, images, videos, even YouTube transcripts

It wraps Perplexica's API and exposes it as an MCP tool. That means your AI assistant can actually search the web on your behalf, cite sources properly, and hand you real information instead of hallucinated facts.

| Mode | When to use | |------|-------------| | speed | Quick fact checks, definitions | | balanced | Everyday research, general questions | | quality | Deep dives, important decisions |

Follow the official installation guide.

git clone https://github.com/ItzCrazyKns/mcp-perplexica.git
cd mcp-perplexica

python -m venv .venv
source .venv/bin/activate  # on Windows: .venv\Scripts\activate
pip install -r requirements.txt

cp .env.example .env

Edit .env with your Perplexica URL:

PERPLEXICA_BASE_URL=http://localhost:3000
LOG_LEVEL=INFO

The default port is 3000, but update PERPLEXICA_BASE_URL if your instance runs elsewhe