#6969th of 107 in Coding agents and AI editors
ralph-orchestrator
Rust CLI that loops coding agent CLIs until a task passes its checks
- Stars
- 3.2k
- License
- MIT
- Last commit
- Oct 2026
- Last release
- Jun 2026
- Language
- Rust
2,374 on 31 Mar 20263,182 on 20 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Ralph Orchestrator implements the Ralph Wiggum technique: it re-runs a coding agent CLI (Claude Code, Codex, Gemini CLI and others) until the agent emits LOOP_COMPLETE or an iteration limit is hit. Work is split across "hats", personas that coordinate through events, and gates such as tests, lint and typecheck reject incomplete output. It also has a Telegram human-in-the-loop bot, an MCP server over stdio, and an alpha web dashboard.
Who it is for: Developers running Claude Code or Codex autonomously on multi-step tasks
Strengths
- Supports 11 agent backends, including Claude Code, Codex, Gemini CLI, Kiro, Amp and Copilot CLI
- Backpressure gates reject incomplete work using tests, lint and typecheck
- Telegram integration lets agents block on questions and accept mid-loop guidance
- Installs via npm, Cargo or a GitHub Releases script
Weaknesses
- Needs a separate agent CLI installed; it does not call models itself
- Web dashboard is alpha with documented rough edges and breaking changes
- Only 5 builtin hat presets: code-assist, debug, research, review, pdd-to-code-assist
- MCP server is scoped to one workspace root, so multi-repo use needs one instance each
What it needs
- no GPU
- Docker + Compose
- Needs Rust 1.75+ (ralph-api), Node.js >= 18 and npm (web dashboard), An agent CLI such as Claude Code, Codex or Gemini CLI, Telegram bot token (optional)
- Models: Claude Code, Kiro, Gemini CLI, Codex, Forge
- port 3000
Also in Coding agents and AI editors
See all 107| Rank | Project | Score |
|---|---|---|
| 1 | claw-codeRust CLI coding agent harness using Anthropic and OpenAI-compatible APIs | 83 out of 100 |
| 2 | ArchonYAML workflow engine that runs AI coding agents through fixed dev steps | 83 out of 100 |
| 3 | Open SWEAgent that turns issues into pull requests, reviews PRs and monitors CI | 81 out of 100 |
| #4 | oh-my-openagentTerminal coding agent with multi-model orchestration and git-backed memory | 80 out of 100 |
| #5 | gooseRust AI agent with desktop app, CLI and API, extended through MCP | 80 out of 100 |