Documentation

Everything you need to get started with Context Pilot — from installation to advanced workflows.

Quick Overview

What is Context Pilot?

Context Pilot is a terminal application (TUI) that wraps around LLM providers to give AI coding assistants structured context management. It provides 58 specialized tools, persistent memory, file edit callbacks, and a modular architecture — all in ~15K lines of pure Rust.

Why does it exist?

Every AI coding assistant — Cursor, Copilot, Claude Code — starts each session blind. No memory of past decisions. No structured view of your codebase. No automatic validation of AI-generated edits. Context Pilot solves all three problems.

Key features

Architecture

Context Pilot is organized as a Cargo workspace with 14 crates. Each module (cp-mod-*) owns its tools, panels, and persistence. The core binary orchestrates the TUI, LLM communication, and tool execution pipeline.

context-pilot/
├── src/              # Core TUI binary (~15K lines)
│   ├── app/          # Application state & event loop
│   ├── modules/      # Module integration layer
│   ├── llms/         # LLM provider implementations
│   └── ui/           # Terminal rendering
├── crates/
│   ├── cp-base/      # Shared types & state
│   ├── cp-mod-files/ # File management module
│   ├── cp-mod-git/   # Git integration
│   ├── cp-mod-memory/# Persistent memory
│   ├── cp-mod-brave/ # Web search (Brave)
│   └── ...           # 9 more modules
└── yamls/            # Embedded config