DEV Community

Dev.Casamia
Dev.Casamia

Posted on

I built a dev tool that "evolves" code with AI — REAP

The Problem

If you've been building with AI agents (like Claude Code), you've probably encountered these problems:

  • Context loss — Start a new session and your context is gone. You end up clinging to long sessions just to avoid losing everything the AI has learned.
  • Stale documentation — You try to persist knowledge in READMEs and CLAUDE.md files, but they quietly go stale as the project moves forward.
  • AI going rogue — Sometimes the AI just ignores your carefully crafted docs and does its own thing anyway.

We're all stuck at the same bottleneck — the context window just isn't enough for long-running projects.

I tried existing tools like spec-kit and superpower — they're decent for one-off feature work, but didn't quite fit for sustained, long-term development.

What I Built

So I built REAP (Recursive Evolutionary Autonomous Pipeline) — an open-source CLI tool inspired by generational evolution in biology.

The idea: AI and humans evolve software across generations.

Genome (Design & Knowledge)
  → Evolution (Generational Progress)
    → Civilization (Source Code)
Enter fullscreen mode Exit fullscreen mode

How It Works

Genome

Your project's design knowledge is managed as a "Genome" — architecture decisions, business rules, conventions, and constraints.

.reap/genome/
├── principles.md      # Architecture principles
├── domain/            # Business rules
├── conventions.md     # Development conventions
└── constraints.md     # Technical constraints
Enter fullscreen mode Exit fullscreen mode

Life Cycle

Each generation follows a five-stage lifecycle:

Objective → Planning → Implementation → Validation → Completion
Enter fullscreen mode Exit fullscreen mode
  • Objective — Define goal, requirements, and acceptance criteria
  • Planning — Break down tasks, choose approach
  • Implementation — Build with AI + human collaboration
  • Validation — Run tests, verify completion
  • Completion — Retrospective + apply Genome changes + archive

Evolution

  • When a generation completes, it gets archived in the lineage, and the next generation picks up new goals.
  • Lessons learned within a generation get folded back into the Genome.
  • Through this iterative pipeline, your source code (the "Civilization") keeps evolving.

Quick Start

# Install
npm install -g @c-d-cc/reap

# Initialize
reap init my-project

# Run a full generation in Claude Code
claude
> /reap.evolve "Implement user authentication"
Enter fullscreen mode Exit fullscreen mode

/reap.evolve runs the entire generation lifecycle — from Objective through Completion — interactively with you.

Links

MIT licensed. Contributions and feedback are welcome!

Top comments (0)