Download
XERJ is an open-source, local AI search engine designed to help developers and AI agents retrieve relevant information from large collections of data. It automatically indexes source code, documents, logs, PDFs, and structured files without requiring manual schema configuration. The engine supports full-text, semantic, vector, and hybrid search to deliver precise results. Its Elasticsearch-compatible API enables integration with existing search clients and tools. Through Model Context Protocol support, AI agents can search indexed repositories and retrieve relevant code passages instead of reading entire files. XERJ also supports retrieval-augmented generation, security auditing, reference coding, and persistent agent memory. Its lightweight architecture enables local deployment across major operating systems.
Features
- Automatic File and Repository Indexing
- Full-Text, Semantic, and Hybrid Search
- Elasticsearch-Compatible Search API
- Model Context Protocol Integration
- Persistent AI Agent Memory
- Code Reference Search and Security Auditing
Project Samples
