LarsCowe/bmalph

Unified AI Development Framework - BMAD phases with Ralph execution loop

What it solves

bmalph streamlines the transition from high-level AI planning to autonomous code implementation. It bundles two distinct systems—BMAD-METHOD for structured planning and Ralph for autonomous execution—into a single CLI tool that integrates with various AI coding assistants (like Claude Code, Cursor, and GitHub Copilot) to ensure that project requirements are systematically analyzed, planned, and then implemented without manual hand-offs.

How it works

bmalph operates in a four-phase workflow:

  1. Initialization: The tool detects or accepts a target AI platform and installs the necessary agent frameworks and instructions into the project directory.
  2. Planning (Phases 1-3): Using the BMAD-METHOD, users work with specialized AI agents (Analyst, PM, Architect) to move from brainstorming and market research to a finalized PRD, technical architecture, and a list of epics and stories.
  3. Implementation Transition: The bmalph implement command converts the planning artifacts (stories and specs) into a format Ralph can understand, creating a prioritized task list (@fix_plan.md).
  4. Autonomous Loop (Phase 4): The bmalph run command starts the Ralph loop, which autonomously picks stories one by one, implements them using Test-Driven Development (TDD), and commits the changes.

Who it’s for

Developers and product managers using AI coding assistants who want a rigorous, agentic workflow to move from a vague idea to a production-ready codebase with minimal manual intervention.

Highlights

  • Multi-Platform Support: Integrates with Claude Code, OpenAI Codex, OpenCode, Cursor, Windsurf, GitHub Copilot, and Aider.
  • Structured Planning: Provides 54 specific command definitions for analysis, planning, and solutioning.
  • Autonomous Execution: Features an autonomous implementation loop with TDD and a live dashboard.
  • Incremental Development: Supports iterative cycles where new epics can be added to BMAD and merged into the existing Ralph implementation plan.
  • Parallel Workers: Ability to run multiple parallel workers in git worktrees via the --swarm flag.

Related

  • Project
    umputun/ralphex

    A standalone CLI tool that orchestrates Claude Code or codex to autonomously execute implementation plans, featuring a multi-phase AI code review pipeline.

  • Project
    frankbria/ralph-claude-code

    An autonomous development loop for Claude Code that iteratively implements project requirements with built-in safeguards against infinite loops and API overuse.

  • Project
    wesammustafa/Claude-Code-Everything-You-Need-to-Know

    A thorough, example‑rich tutorial for Anthropic’s Claude Code CLI, covering installation, prompt engineering, and advanced extensions (Skills, Hooks, Sub‑agents, Workflows, MCP) to let developers use Claude for repo‑wide code analysis, editing, testing and automation.

  • Project
    intelligencedev/manifold

    An experimental platform for long-horizon workflow automation that enables teams of AI specialists to collaborate on complex, multi-step objectives.

  • Project
    nrslib/takt

    A CLI tool that orchestrates AI coding agents into repeatable, YAML-defined development workflows with isolated worktrees and explicit review loops.