LLocalSearch

Normal
GitHub Go Apache-2.0

Description

A completely locally running search aggregator using LLM agents. Users can ask questions and the system uses a chain of LLMs to find answers without any external API keys.

Key Features

  • Completely local operation with no API keys required — full privacy protection
  • Recursive LLM agent chain for multi-step internet search and reasoning
  • Live logs and source links providing transparency into agent reasoning
  • Supports follow-up questions for iterative research
  • Mobile-friendly responsive design with dark and light mode

Use Cases

💡 Privacy-respecting web search without relying on commercial search engines
💡 Research tasks requiring multi-source information aggregation
💡 Educational exploration of how LLM agents use tools recursively
💡 Local knowledge base queries using self-hosted language models

Categories

Quick Start

Clone the repo, run `docker-compose up -d`. Ensure Ollama is running with a supported model. Configure `.env` if Ollama runs on a different device. Access the web UI and start asking questions.

Related Projects