Search everything...

Stats

Actions

Available In

Help us improve

Share bugs, ideas, or general feedback.

quorum - Claude Code Plugin | ClaudePluginHub

Plugin

quorum

Name: quorum
Author: nexusentis

By nexusentis

Multi-agent quorum workflow — fans out tasks to Claude, Codex, Copilot, Cursor, and Gemini in parallel, then reconciles their outputs into a high-confidence recommendation. Use for planning, code review, implementation validation, bug diagnosis, and architecture decisions.

npx claudepluginhub nexusentis/quorum --plugin quorum

Popularity

Stars

Above avg

Med: 0·Avg: 274

Installs

Med: 0·Avg: 1

What's Inside

Agents5

claude-agent

/claude-agent

Directly analyzes code and answers prompts using filesystem exploration tools. Used exclusively by the /quorum skill for parallel fan-out alongside the relay agents.

codex-agent

/codex-agent

Dispatches a prompt to OpenAI Codex CLI via the quorum-codex MCP server and returns the raw structured result. Used exclusively by the /quorum skill for parallel fan-out.

copilot-agent

/copilot-agent

Dispatches a prompt to GitHub Copilot CLI (ACP mode with plain fallback) via the quorum-copilot MCP server and returns the raw structured result. Used exclusively by the /quorum skill for parallel fan-out.

cursor-agent

/cursor-agent

Dispatches a prompt to Cursor CLI (Composer 2) via the quorum-cursor MCP server. Returns the raw structured result. Used exclusively by the /quorum skill for parallel fan-out.

gemini-agent

/gemini-agent

Dispatches a prompt to Google Gemini CLI (gemini-3.1-pro) via the quorum-gemini MCP server and returns the raw structured result. Used exclusively by the /quorum skill for parallel fan-out.

Skills2

agent

/agent

Run a single quorum agent (claude, codex, copilot, cursor, or gemini) with a prompt. Usage - /quorum:agent <agent> <prompt>

run

/run

Fan out a task to Claude, Codex, Copilot, Cursor, and Gemini in parallel, then reconcile their outputs into a high-confidence recommendation. Use for planning, code review, implementation validation, bug diagnosis, and architecture decisions — not for quick changes.

MCP Servers5

quorum-codex

admin

Cross-check: OpenAI Codex CLI wrapper (gpt-5.5-codex)

quorum-copilot

admin

Cross-check: GitHub Copilot CLI wrapper (ACP + fallback)

quorum-cursor

admin

Cross-check: Cursor CLI wrapper (Composer 2)

quorum-gemini

admin

Cross-check: Google Gemini CLI wrapper (gemini-3.1-pro)

quorum

admin

Quorum: fan out to all agents in parallel and return collected results

Stats

Version1.1.3

ReleasedApr 27, 2026

LanguageTypeScript

Stars1

MaintenanceExcellent

LicenseMIT

Last CommitApr 27, 2026

AddedMar 21, 2026

Actions

View on GitHub View README Plugin Marketplace JSON

Own this plugin?

Verify ownership to unlock analytics, metadata editing, and a verified badge.

Available In

quorum-marketplace1

Safety Signals

Critical

Admin access level

Server config contains admin-level keywords

Caution

Uses power tools

Uses Bash, Write, or Edit tools

Help us improve

Share bugs, ideas, or general feedback.

README

Multi-agent MCP server that fans out prompts to Claude, Codex, Copilot, Cursor, and Gemini in parallel, then reconciles their outputs into a single high-confidence recommendation.

Quickstart (Claude Code)

Install as a Claude Code plugin:

/plugin marketplace add nexusentis/quorum
/plugin install quorum@quorum-marketplace

Then install the agent CLIs you want to use (see Prerequisites), and run:

/quorum:run review this repo

Two usage paths

Quorum works in two distinct ways depending on your editor:

Claude Code — full orchestration

Claude Code gets the complete experience: skills, relay agents, and blind judge reconciliation.

/quorum:run <prompt> — Fan out to all enabled agents, deduplicate findings, run blind judge scoring, and present a synthesized recommendation. Supports modes: review, review-codebase, architecture, implement, diagnose.
/quorum:agent <agent> <prompt> — Test a single agent and get the raw result back. Useful for debugging.

Claude participates as both an agent (analyzing code directly) and the judge (scoring anonymized outputs). Agent identities are hidden during scoring to prevent bias.

Any MCP client — direct tool calls

Cursor, VS Code, Windsurf, and any other MCP-compatible client get access to the MCP tools directly. The client's own LLM decides how to use the results.

codex_query / copilot_query / cursor_query / gemini_query — Query a single agent
quorum_query — Fan out to all enabled agents in parallel and return collected results

No skills, no relay agents, no blind judge — your editor's LLM handles interpretation.

Installation

Claude Code (recommended)

Install directly as a plugin:

/plugin marketplace add nexusentis/quorum
/plugin install quorum@quorum-marketplace

This sets up everything automatically — MCP servers, relay agents, skills, and permissions. Restart Claude Code after installing.

What gets installed

.mcp.json — 5 MCP servers (4 individual agents + combined quorum)
settings.json — Auto-approve permissions for all agent CLIs and MCP tools
agents/ — 5 relay agent definitions for parallel fan-out via the Task tool
skills/ — /quorum:run (full orchestration) and /quorum:agent (single agent)
quorum.config.json — Enable/disable individual agents

Manual setup (from source)

Clone the repo and install dependencies:

git clone https://github.com/nexusentis/quorum.git
cd quorum/mcp-servers && npm install
# Restart Claude Code to pick up the MCP servers

.mcp.json

{
  "mcpServers": {
    "quorum-codex": {
      "command": "./mcp-servers/run.sh",
      "args": ["codex-server.ts"],
      "description": "Cross-check: OpenAI Codex CLI wrapper (gpt-5.5-codex)"
    },
    "quorum-copilot": {
      "command": "./mcp-servers/run.sh",
      "args": ["copilot-server.ts"],
      "description": "Cross-check: GitHub Copilot CLI wrapper (ACP + fallback)"
    },
    "quorum-cursor": {
      "command": "./mcp-servers/run.sh",
      "args": ["cursor-server.ts"],
      "description": "Cross-check: Cursor CLI wrapper (Composer 2)"
    },
    "quorum-gemini": {
      "command": "./mcp-servers/run.sh",
      "args": ["gemini-server.ts"],
      "description": "Cross-check: Google Gemini CLI wrapper (gemini-3.1-pro)"
    },
    "quorum": {
      "command": "./mcp-servers/run.sh",
      "args": ["quorum-server.ts"],
      "description": "Quorum: fan out to all agents in parallel and return collected results"
    }
  }
}

settings.json

{
  "permissions": {
    "allow": [
      "Bash(codex exec --full-auto --sandbox read-only *)",
      "Bash(copilot -p * --silent *)",
      "Bash(copilot --acp)",
      "Bash(agent -p --force --trust *)",
      "Bash(gemini -p * --yolo *)",
      "mcp__quorum-codex__codex_query",
      "mcp__quorum-copilot__copilot_query",
      "mcp__quorum-cursor__cursor_query",
      "mcp__quorum-gemini__gemini_query",
      "mcp__quorum__quorum_query"
    ]
  }
}

Cursor

cd your-project
npx quorum-agents --setup cursor

This creates .cursor/mcp.json:

View full README on GitHub

Help us improve

Find plugins for your project

Help us improve

quorum

Popularity

What's Inside

Help us improve

Health & Quality

Confidence

README

Quickstart (Claude Code)

Two usage paths

Claude Code — full orchestration

Any MCP client — direct tool calls

Installation

Claude Code (recommended)

Cursor

Similar Plugins

owlex

agent-tower

claude-council

consensus

deliberation

llm-council-plugin

Help us improve

Quickstart (Claude Code)

Two usage paths

Claude Code — full orchestration

Any MCP client — direct tool calls

Installation

Claude Code (recommended)

Cursor