lxyer/multi-agent-collaboration-system
Terminal-native cockpit for AI-first software engineering — protocol-driven sidecar across Claude Code, Codex CLI, OMX, OMCC and beyond.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit lxyer/multi-agent-collaboration-system/.claude ./rig-multi-agent-collaboration-system # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/before-dev .claude/skills/before-dev $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/brainstorm .claude/skills/brainstorm $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/break-loop .claude/skills/break-loop $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/check-cross-layer .claude/skills/check-cross-layer $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/check .claude/skills/check $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/create-command .claude/skills/create-command $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/finish-work .claude/skills/finish-work $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/improve-ut .claude/skills/improve-ut $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/integrate-skill .claude/skills/integrate-skill $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/onboard .claude/skills/onboard $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/record-session .claude/skills/record-session $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/start .claude/skills/start $ npx degit lxyer/multi-agent-collaboration-system/.agents/skills/update-spec .claude/skills/update-spec $ npx degit lxyer/multi-agent-collaboration-system/.codex/skills/parallel .claude/skills/parallel
$ curl -fsSL --create-dirs -o .claude/agents/check.md https://raw.githubusercontent.com/lxyer/multi-agent-collaboration-system/main/.claude/agents/check.md $ curl -fsSL --create-dirs -o .claude/agents/debug.md https://raw.githubusercontent.com/lxyer/multi-agent-collaboration-system/main/.claude/agents/debug.md $ curl -fsSL --create-dirs -o .claude/agents/dispatch.md https://raw.githubusercontent.com/lxyer/multi-agent-collaboration-system/main/.claude/agents/dispatch.md $ curl -fsSL --create-dirs -o .claude/agents/implement.md https://raw.githubusercontent.com/lxyer/multi-agent-collaboration-system/main/.claude/agents/implement.md $ curl -fsSL --create-dirs -o .claude/agents/plan.md https://raw.githubusercontent.com/lxyer/multi-agent-collaboration-system/main/.claude/agents/plan.md $ curl -fsSL --create-dirs -o .claude/agents/research.md https://raw.githubusercontent.com/lxyer/multi-agent-collaboration-system/main/.claude/agents/research.md
Merge into .claude/settings.json (project) or ~/.claude/settings.json (user). Hook commands reference scripts in the source repo — copy those too.
{
"hooks": {
"PreToolUse": [
{
"matcher": "Task",
"hooks": [
{
"type": "command",
"command": "python3 .claude/hooks/inject-subagent-context.py"
}
]
},
{
"matcher": "Agent",
"hooks": [
{
"type": "command",
"command": "python3 .claude/hooks/inject-subagent-context.py"
}
]
}
]
}
} Skills (14)
before-devbrainstormbreak-loopcheck-cross-layercheckcreate-commandfinish-workimprove-utintegrate-skillonboardrecord-sessionstartupdate-specparallel
Subagents (6)
| check model: opus | | |
| debug model: opus | | |
| dispatch model: opus | | |
| implement model: opus | | |
| plan model: opus | | |
| research model: opus | | |
Hooks (6)
| event | matcher | runs |
|---|---|---|
| SessionStart | startup | python3 .claude/hooks/session-start.py |
| SessionStart | clear | python3 .claude/hooks/session-start.py |
| SessionStart | compact | python3 .claude/hooks/session-start.py |
| PreToolUse | Task | python3 .claude/hooks/inject-subagent-context.py |
| PreToolUse | Agent | python3 .claude/hooks/inject-subagent-context.py |
| SubagentStop | check | python3 .claude/hooks/ralph-loop.py |
Slash commands (13)
/trellis/before-dev/trellis/brainstorm/trellis/break-loop/trellis/check-cross-layer/trellis/check/trellis/create-command/trellis/finish-work/trellis/integrate-skill/trellis/onboard/trellis/parallel/trellis/record-session/trellis/start/trellis/update-spec
Cursor rules (1)
macs-protocol · always
Similar rigs
arunpaul-H/coding-agent-harness
Your coding agent's rules, enforced by code instead of prompts. Hooks block bad actions before they run; a committed baseline turns a deleted or skipped test into a red build. Zero dependencies. Hooks for Claude Code, gates for any CI.
Automator 3.1k tok ·
slowernet/dietpowers
An opinionated test-first development flow plugin for Claude Code, inspired by Superpowers [WIP]
Skill Collector 2.1k tok ·
serafim-builds/PersonaOS
An open-source AI that quietly remembers who you are — in plain markdown you fully own. Works with Claude Code, Codex CLI, opencode, and any other agent that reads AGENTS.md.
Skill Collector 13.4k tok ·
itsOmidKarami/kraft
A local orchestrator that takes your coding agent from spec to pull request, stopping only when a decision is yours.
Skill Collector 7.9k tok ·