umputun/ralphex

Extended Ralph loop for autonomous AI-driven plan execution

What it solves

ralphex is a CLI tool designed to automate the execution of complex software implementation plans using AI coding agents like Claude Code or codex. It eliminates the need for manual "babysitting" of interactive agents and prevents the quality degradation that occurs when a single long AI session fills up with too much context. By executing each task in a fresh session, it keeps the AI model sharp and focused.

How it works

ralphex orchestrates a multi-phase pipeline to turn a markdown-based plan into completed code:

  1. Task Execution: It reads a plan file, executes incomplete tasks one by one using an AI agent, and runs validation commands (like tests or linters) after each step.
  2. First Code Review: It launches five specialized agents (quality, implementation, testing, simplification, and documentation) in parallel to find and fix issues.
  3. External Review: It optionally uses an external tool (such as OpenAI Codex) to provide an independent cross-model review.
  4. Second Code Review: A final pass by two agents focuses on critical and major issues to ensure no regressions occurred.
  5. Finalize: An optional step that can rebase and squash commits to clean up the git history.

Who it’s for

Developers who want to automate feature implementation and code reviews using AI agents without needing an IDE plugin or cloud service, and those who prefer a structured, plan-driven approach to AI-assisted coding.

Highlights

  • Autonomous Pipeline: Full automation from plan creation to final commit.
  • Multi-Agent Review: A sophisticated pipeline using multiple specialized agents and optional cross-model verification.
  • Worktree Isolation: Ability to run multiple plans in parallel using git worktrees.
  • Flexible Executors: Supports both Claude Code and codex as the primary execution engine.
  • Real-time Monitoring: Includes a streaming terminal output and an optional web dashboard for tracking progress.

Related

  • Project
    robzilla1738/supergoal

    An autonomous planning and execution framework for Claude Code and Codex CLI that transforms complex tasks into a single, self-healing autonomous run.

  • Project
    chaseai-yt/claudex-loop

    Claudex Loop is a multi-agent workflow that improves AI coding reliability by requiring independent AI models to review plans and inspect code generated by a primary builder.

  • Project
    umputun/cc-thingz

    A marketplace of Claude Code plugins that add brainstorming, PR review, planning, release automation, thinking aids, skill‑evaluation hooks, and workflow helpers. Install via `/plugin marketplace add umputun/cc-thingz` and pick the plugins you need.

  • Project
    AnandChowdhary/continuous-claude

    A CLI tool that runs AI coding agents in a continuous loop to autonomously create, test, and merge pull requests for large-scale software projects.

  • Project
    claesbackman/AI-research-feedback

    A collection of Claude Code “skills” that let the Anthropic Claude assistant review papers, grant proposals, pre‑analysis plans, and analysis code, turn papers into policy briefs or slide decks, and convert PDFs to markdown. Install as `SKILL.md` files, invoke with slash commands (e.g., `/review-paper`), and receive structured markdown reports or HTML outputs.