ragnos-labs

Merge 2 Main — Git skill for Claude Code

Git community

Multi-agent coordination framework for Claude Code and Codex.

How to install Merge 2 Main

This entry records only its repository, not the path inside it, so there is no exact command to give. Open ragnos-labs/merge-2-main and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Merge 2 Main does

Multi-agent coordination framework for Claude Code and Codex.

Alternatives in Git

Epic Sync — Sync epic issue bodies, labels, and local coordination snapshots from GitHub 243.5k ★
Claude Relay Service — GitHub Repo stars Self-built Claude code mirror service supporting multi-account switching Relay service 9.9k ★
Merge Command — Finalize worktree work: update docs, commit, merge to main, clean up 1.3k ★

README

merge-2-main by Mr. CLI

[![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE) [![Discussions](https://img.shields.io/github/discussions/ragnos-labs/merge-2-main)](https://github.com/ragnos-labs/merge-2-main/discussions)

**A living field guide for multi-agent coding, review, and release discipline.**

Patterns, guides, references, and templates for orchestrating 2 to 30+ AI agents on real software engineering tasks. This repo is documentation-only by design: a public methodology layer, not a framework, SDK, or hosted service.

Agent work is chaotic right now. This is the field guide I wished existed when I started.

Patterns battle-tested across multiple production agent fleets, distilled into a methodology layer anyone can adopt or argue with.

Start Here

Three minutes to a working pattern:

Open the Decision Tree. Pick a pattern.
Open the matching runtime. Pick a surface.
Grab a template. Run it.

Or skim the [Quick Win](#quick-win-patchwork-in-90-seconds) below to see a single-agent pattern in action without leaving this page.

What This Repo Is

A source-backed methodology library for agentic software work. The goal is to document the coordination patterns, review gates, audit hooks, and runtime adapters that help teams move quickly without losing traceability.

You get:

coordination patterns from single-agent work to large swarms
a git-isolation layer for parallel workstreams
cross-cutting guides for planning, tests, checkpoints, release gates, and artifacts
references for anti-patterns, handoffs, sourcing, and schemas
worked examples showing full end-to-end workflows
ready-to-use templates for prompts, configs, and sprint artifacts

Works with [Claude Code](https://code.claude.com/docs/en), [OpenAI Codex](https://openai.com/index/introducing-codex/), and [OpenClaw](https://openclaw.im/docs/start/set