Cognito AI Search provides a private, self-hosted gateway to information, combining a local AI assistant with a SearXNG metasearch engine. It targets privacy-conscious individuals seeking control over their digital footprint, offering an alternative to commercial search engines and cloud-based AI.

How It Works

The application integrates with local AI models via Ollama for private conversational AI and a self-hosted SearXNG instance for anonymous web searches. This dual approach ensures user data remains on their machine, enabling offline AI interaction and unbiased search results aggregated from multiple sources without tracking.

Quick Start & Requirements

• Install/Run: docker run -d -p 3000:3000 -e OLLAMA_API_URL="http://YOUR_OLLAMA_HOST:11434" -e SEARXNG_API_URL="http://YOUR_SEARXNG_HOST:8888" --name cognito-ai-search kekepower/cognito-ai-search:latest
• Prerequisites: Ollama instance, SearXNG instance.
• Configuration: Requires setting OLLAMA_API_URL and SEARXNG_API_URL environment variables.
• Documentation: HOWTO

Highlighted Details

• Features a "Holographic Shard Design" UI with neon accents and glass morphism.
• Claims 50% faster performance with optimized dependencies and reduced component count.
• Offers enhanced search with AI-focused suggestions and professional PDF export with LaTeX support.
• Supports dual light and dark themes with advanced caching and clean error handling.

Maintenance & Community

• Community engagement encouraged via GitHub starring, issue reporting, and feature suggestions.
• Links to GitHub for contributions and issue tracking.

Licensing & Compatibility

• Licensed under the MIT License.
• Permissive license suitable for commercial use and integration with closed-source applications.

Limitations & Caveats

The project requires separate, self-hosted instances of Ollama and SearXNG to be functional, adding to the setup complexity beyond just running the Docker container.