open-ralph-wiggum
Type `ralph "prompt"` to start open code in a ralph loop. Also supports a prompt file & status check. Open Code, Claude Code, Codex, Copilot
Open Ralph Wiggum is a command-line tool that creates autonomous iterative coding loops for AI agents including Claude Code, Codex, Copilot CLI, Cursor Agent, Qwen Code, and OpenCode. It implements the Ralph Wiggum technique where an AI coding agent receives the same prompt repeatedly until task completion, with each iteration allowing the agent to see and self-correct its previous work through file changes and git history. The tool supports multi-agent switching, task management, live monitoring, and mid-loop context injection, making it ideal for long-running development tasks that benefit from automated iteration and self-correction. Developers can set maximum iterations, define completion signals, and monitor progress from separate terminals while the agent works autonomously.
Key Features
Use Cases
- 01Building REST APIs with CRUD operations and automated test verification across multiple iterations
- 02Refactoring codebases with automatic validation through linters, type checkers, and test suites
- 03Implementing complex features broken into subtasks with automatic progression through Tasks Mode
- 04Comparing performance of different AI coding models through agent rotation on the same task
- 05Greenfield project development where the agent can work autonomously until all acceptance criteria pass
- 06Debugging and fixing failing tests iteratively until the entire test suite passes
Related Plugins
View moreandrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
claude-mem
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
Understand-Anything
Graphs that teach > graphs that impress. Turn any code into an interactive knowledge graph you can explore, search, and ask questions about. Works with Claude Code, Codex, Cursor, Copilot, Gemini CLI, and more.
rtk
CLI proxy that reduces LLM token consumption by 60-90% on common dev commands. Single Rust binary, zero dependencies
open-ralph-wiggum — FAQ
What is Open Ralph Wiggum and how does it work?+
Open Ralph Wiggum is a CLI tool that wraps AI coding agents (Claude Code, Codex, Copilot CLI, Cursor Agent, Qwen Code, or OpenCode) in autonomous iterative loops. It sends the same prompt repeatedly until the agent completes the task, with each iteration allowing the agent to see and correct its previous work through file system and git history changes.
How do I install Open Ralph Wiggum?+
Install via npm with 'npm install -g @th0rgal/ralph-wiggum' after installing the Bun runtime and at least one AI coding agent CLI (Claude Code, Codex, Copilot CLI, Cursor Agent, Qwen Code, or OpenCode). You can also install from source by cloning the repository and running the install script.
Which AI coding assistants work with Open Ralph Wiggum?+
It works with Claude Code, OpenAI Codex, GitHub Copilot CLI, Cursor Agent, Qwen Code, and OpenCode. Switch between agents using the --agent flag, for example: 'ralph "task" --agent claude-code --max-iterations 10'.
What are the prerequisites and do I need API keys?+
You need the Bun runtime and at least one AI coding agent CLI installed. API key requirements depend on your chosen agent: Copilot CLI requires GitHub authentication via GH_TOKEN or GITHUB_TOKEN, Cursor Agent may need CURSOR_API_KEY for headless environments, and other agents require their respective authentication setup.
Is Open Ralph Wiggum free to use?+
Yes, Open Ralph Wiggum itself is free and open-source under the MIT license. However, the underlying AI agents may have their own costs (like Copilot subscription fees) or API usage charges depending on which agent and model you use.
How do I monitor an active Ralph loop without stopping it?+
Run 'ralph --status' from another terminal to see the current iteration, elapsed time, iteration history, struggle indicators, and active tasks. You can also inject guidance mid-loop using 'ralph --add-context "your hint here"' without interrupting the running process.
How do I install open-ralph-wiggum?+
Open the source repository on GitHub and follow its README. open-ralph-wiggum is a plugin — MCP Agents Market links you directly to the official repo.
Is open-ralph-wiggum free?+
open-ralph-wiggum is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.