Code Context Engine
A local codebase index that lets AI coding agents search instead of re-reading files, cutting input tokens by a reproducibly benchmarked 94 percent across Claude Code, Codex, Copilot, Cursor and Gemini CLI.
Code Context Engine indexes a codebase locally so questions hit the index instead of causing the agent to re-read files, sharing one index across multiple editors for teams that mix Claude Code, Cursor, VS Code and Gemini CLI. It keeps everything local with zero cloud indexing, tracks dollar-amount savings rather than estimates, and preserves decisions and context across restarts.
Setup is a single command that installs, indexes and configures the tool, auto-detecting the editor and writing its config, with an Ollama auto-detection path when no other embedding option is configured. An agent flag targets a specific client such as Claude, Codex, Copilot or Kiro, and a plugin flag generates a portable plugin directory that launches the engine on demand without requiring users to pre-install the Python package. It requires Python 3.11 or newer and a C compiler for tree-sitter grammars, and is tested on macOS, Linux and Windows.