Vexp

About

vexp provides a local context engine that supercharges AI coding agents. By pre-indexing your code into a dependency graph, vexp delivers only the necessary context through MCP. This prevents Claude Code, Cursor, and similar agents from inefficiently using tokens on unnecessary file searches. In FastAPI benchmark tests involving 42 runs, vexp achieved a 58% reduction in cost, a 63% decrease in output tokens, and a 90% drop in tool calls. It operates entirely locally, eliminating the need for cloud services or accounts. A free usage level is offered.