Code2Prompt

Code2Prompt is a context engineering tool that ingests your codebase and transforms it into structured, AI-ready prompts. Built in Rust for high performance, it enables developers to convert any repository into meaningful context following the Goal + Format + Context framework, making it easier to get accurate, context-aware responses from AI models.

Key Features:

• High-Performance Rust Engine - Written in Rust for speed and efficiency, capable of handling large codebases with minimal resource usage while maintaining fast processing times.

• Handlebars-Powered Templates - Offers customizable prompt generation using Handlebars templates, giving developers full control over the output format and structure of generated prompts.

• Smart Filtering with Glob Patterns - Supports include/exclude patterns with glob matching for precise code selection, allowing you to query LLMs without extra noise and reduce hallucination.

• Multi-Format Export - Exports structured prompts in JSON, Markdown, or XML with different formatting options to suit various workflows and AI model requirements.

• Git Integration - Includes Git diff and log extraction for better context, making it easier to understand code changes over time and provide relevant history to AI models.

• Multiple Interfaces - Available as a CLI tool for terminal usage, a Python SDK for programmatic integration, and an MCP (Model Context Protocol) server for AI agent workflows.

• LLM Compatibility - Works with all major AI models including Claude, GPT, Gemini, Mistral, DeepSeek, Qwen, Grok, and local models via Ollama.

Getting Started:

To use Code2Prompt, install it via the CLI and run a simple command like code2prompt . in your project directory. Use flags like --include "*.js,*.html" and --exclude "node_modules/" to filter relevant files. The tool generates a structured prompt containing your project tree and file contents, ready to paste into any AI assistant. For Python applications, install the code2prompt-rs package from PyPI. For AI agents, integrate the MCP server to provide rich codebase context automatically.