Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
14 changes: 14 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -150,6 +150,20 @@ its own git worktree. Nothing runs autonomously — you approve every build.

## Quickstart

**Which install should I use?**

| You are… | Use | How |
|---|---|---|
| On **Claude Code** and want typed `/loop:*` commands | **Plugin** (C) | `/plugin marketplace add anshulixyz/multi-agent-loop-kit` then `/plugin install loop@multi-agent-loop-kit` |
| On **Claude Code** and prefer the Agent Skills workflow | **Skill** (D) | clone into `~/.claude/skills/multi-agent-loop`, then ask Claude to "set up a multi-agent loop here" |
| Using **Cursor / Codex / another agent** (no Claude Code) | **Clone-install** (A/B) | run `scripts/install-into-repo.sh` and point your agent at `prompts/setup-interview.md` |
| Just **trying it out** | **New repo** (A) | clone the kit and explore on a throwaway repo |

The plugin and skill are two faces of the same kit on Claude Code — pick by taste
(namespaced commands vs one model-invocable skill). All paths wrap the same scripts.
The skill/plugin run **locally**; they are not for the claude.ai/Desktop uploaded-skill
sandbox (no local git worktrees there). Details for each below.

### A. New / empty repo (recommended starting point)

Easiest way to learn the kit is on a fresh repo where nothing can conflict.
Expand Down
Loading