Repo Prompt

Repo Prompt is a macOS-native application that serves as a Context IDE for AI agents. It automates context curation from a codebase, provides deep analysis, and enables multi-agent orchestration. The tool helps developers move from manually prompting AI models to orchestrating complex engineering work.

Key Features

• Context Builder: Automatically explores a codebase, selects relevant files within a token budget, and provides curated context to analysis models.
• Multi-Agent Orchestration: Plan tasks with a reasoning model, then dispatch sub-agents across different providers (e.g., Claude Code, Codex CLI, Gemini CLI) to implement work. Supports both sequential and parallel execution.
• Agent Mode: A native interface to run CLI agents with full codebase context and token-efficient MCP tools.
• Workflows: Includes slash commands like /rp-orchestrate, /rp-review, /rp-refactor, and /rp-investigate for common development tasks.
• MCP Server: Offers 15+ token-efficient MCP tools accessible from any MCP-compatible client.
• CodeMaps & Token Efficiency: Uses tree-sitter for function signatures, allowing agents to see more files efficiently.
• Oracle Collaboration: Agents can consult powerful reasoning models mid-session using the full repo for answers.
• Repo Bench: A community-driven benchmark to measure AI coding performance across models.

Who It's For

Repo Prompt is designed for developers using AI coding agents, particularly those working with tools like Claude Code, Cursor, OpenCode, and Gemini CLI. It also supports personal agents like OpenClaw.

Pricing

Repo Prompt offers a Free tier with core features and a Pro subscription (monthly or yearly) that unlocks Context Builder, CodeMaps, Agent Mode, Orchestration, MCP Server integration, and usage on up to 3 machines. Team licenses are also available.