repoprompt/repoprompt-ce

What it solves

RepoPrompt CE addresses the difficulty of providing AI coding agents with the precise codebase context they need to be effective. Instead of dumping entire repositories into a prompt, it allows users to curate focused, reviewable context—including file contents, repository structures, and Git diffs—to stay within token budgets and improve model accuracy.

How it works

It is a native macOS application that acts as an agent orchestrator and context engineering tool. It uses a bundled Model Context Protocol (MCP) server to allow external MCP-compatible clients and CLI agents to search repositories, inspect files, and curate context. The app provides a "Compose inspector" for reviewing and refining the final prompt before it is sent to an AI model.

Who it’s for

Developers using AI coding agents and CLI-based AI tools who want a visual interface to manage the codebase context being fed into their LLMs on macOS.

Highlights

• Context Engineering: Combines file trees, line slices, CodeMaps, and Git diffs into dense prompts.
• Agent Orchestration: Coordinates CLI-backed coding agents directly from the macOS app.
• MCP Integration: Includes an MCP server to connect external tools to the repository context.
• Multi-root Workspaces: Supports working across multiple related repositories or documentation folders simultaneously.
• Reviewable Handoffs: Allows users to inspect and refine context before sending it to another agent or model.