# Vexor

A semantic search engine for files and code that builds a reusable index and lets a developer or an AI coding assistant find a file by describing what it does, through a CLI, a Python API or an MCP server.

Vexor auto-indexes a directory on first search and supports configurable embedding and reranking providers, exposing the same core functionality through three interfaces built on one shared engine: a command-line tool for interactive searches such as vexor search with an explicit path and result count, a Python API with index and search functions for programmatic use, and an MCP server for AI coding assistants doing autonomous semantic file discovery.

A guided setup wizard configures the tool the first time it runs, or can be invoked directly, and configuration-management, MCP, help and version commands run without needing that setup first. It installs from a standalone binary with no Python required, or with pip, pipx or uv, and is designed for both human developers and AI coding assistants working in autonomous agent workflows.

Website: https://github.com/scarletkc/vexor