Amicus Open Beta macOS Windows

Every model.
One verdict.

LLM Council. Parallel window.

Hand Claude a doc, a diff, or a design and say council review this: Amicus routes it through models from different families, has them cross-review each other anonymously, and a non-Claude chair synthesizes the verdict. Or skip the ceremony — fork a single conversation to Gemini, GPT, DeepSeek, or any other model, work in parallel with full context, and fold the result back.

$npm install -g amicuscopy

Docs

Any model via OpenRouter

Quick Start

Install. Configure. Council.

One install delivers both skills, the CLI, and the MCP server. No manual registration required.

# Install
npm install -g amicus

# Configure API keys and default model
amicus setup

# Then in Claude Code or Cowork, just say:
"council review this"

> Amicus routes your doc through 3 models, cross-reviews, and presents a tiered verdict.
> You make the accept / deny calls.

How It Works

Council. Or Fork.
You Choose the Depth.

A full multi-model council for structured review, or a quick fork to one other model. Both share your full Claude context automatically.

01 · Council

Say council review this. Independent reviews → anonymous cross-review → a non-Claude chair verdict. You accept or deny tiered findings.

02 · Fork & Work

Or fork to a single model — Gemini, GPT, DeepSeek, any. It starts with your full Claude context. Work alongside it in a real window.

03 · Fold

Click Fold and a structured summary flows back into Claude. Findings, recommendations, decisions. No context switching, no noise.

Parallel Subagents

Spawn Subagents.
Claude Orchestrates.

Amicus can also run headlessly as parallel subagents inside Claude Code or Cowork. Spawn multiple models at once. Each runs independently on its task. All results fold back into Claude.

Claude as orchestrator. Any model as a specialist.

From inside Claude Code or Cowork, ask Claude to spawn headless Amicus instances using the MCP tool. Each subagent gets your full session context, works autonomously on its assigned task, and returns a structured summary. No window switching, no prompting each model yourself.

Run multiple Amicus instances simultaneously to split work across specialized models. One reviews architecture. Another audits security. A third generates tests. Claude collects every summary and synthesizes the results back into your main context.

Native MCP tool works inside Claude Code and Cowork

Each subagent gets full conversation context automatically

Runs autonomously with a configurable timeout

Structured summaries fold back into Claude's context

Conflict detection prevents subagents from overwriting each other

Automatic Context Sharing

Your Claude Context.
Any Model.

Amicus reads your active Claude Code session and passes it to whichever model you choose. No export, no copy-paste, no starting from scratch.

The only tool that bridges Claude Code sessions to other AI models. Every Amicus session starts with everything Claude already knows.

Conversation history File changes Tool calls Error output

Works With

Claude Code & Cowork.
Two Interfaces, One Tool.

Amicus integrates natively with both Claude Code and Claude Cowork. Install once, use everywhere.

CLI

Claude Code

Use the amicus command directly in your terminal alongside Claude Code. Desktop app and CLI both supported.

          amicus start --model gemini --prompt "Review auth"
        

MCP Server

Claude Cowork

Amicus registers as an MCP server automatically. Cowork agents can spawn Amicus instances natively from their sandbox.

          Tool: amicus_start model: gemini
        

See It In Action

One Command.
Full Context.

# Fact-check Claude's approach with Gemini
amicus start --model gemini --prompt "Verify the auth refactor"

# Deep-dive without polluting your main context
amicus start --model gemini-pro --prompt "Analyze the codebase"

# Autonomous background task
amicus start --model gemini --no-ui --prompt "Generate tests"

> Started Amicus gk7x · google/gemini-3-flash-preview
> Context: 42 turns, 18.2k tokens
> Running autonomously... (15m timeout)

Use Cases

Built for How
You Actually Work.

Every Amicus session shares your context automatically. Just pick a model and get to work.

Fact-Check

Claude proposed an architecture? Send it to Gemini for a second opinion. Catch bad assumptions before they become bugs.

Debug

Stuck on a bug? Bring in a different model for a second look. A new perspective often catches what you've stopped seeing.

Brainstorm

Get three different models thinking about the same problem in parallel. Claude collects and synthesizes the best ideas from each.

Fresh Eyes

Deep in a session and losing perspective? Bring in a fresh model. It sees everything you've built, without the tunnel vision.

Your Tools Follow You

Every MCP server configured in Claude Code is automatically discovered and available inside Amicus. Repomix, GitHub, Slack, whatever tools you use. Zero extra setup.

Compatible Models

Any Model.
Your Keys.

Use your existing API keys directly, or connect everything through OpenRouter with a single key.

Google Gemini 3.5 Flash, 3.1 Pro

OpenAI GPT-5.5, o3, o4-mini

Anthropic Claude Opus, Sonnet, Haiku

xAI Grok 4.20, Grok 4.3

Meta Llama 4 Scout, Maverick

DeepSeek V4 Pro, V4 Flash, R1

OpenRouter

200+ models from every provider. One API key for everything.

Direct API Keys

Already have a Google AI, OpenAI, Anthropic, or DeepSeek key? Use it directly. No middleman, no extra accounts.

OpenRouter (Recommended)

One key, every model. Automatic fallback, unified billing. Run amicus setup to configure.

Features

Everything You Need.
Nothing You Don't.

Multi-Model Council

Independent reviews → anonymous cross-review → non-Claude chair verdict. Blind spots surface when models from different families disagree.

Fork / Work / Fold

Spin up one other model in a real window with full context, work alongside it, and fold a structured summary back when you're done.

Fanout Parallel Waves

Run N models on the same prompt headlessly in one wave. Every leg gets the same briefing; results arrive as a single JSON wave document.

Live Model Catalog

No frozen model table. Aliases and validation resolve against a live catalog fetched from provider APIs — model names stay current without a code change.

Windows First-Class

Developed and tested on Windows 11, no WSL required. Full unit suite runs green on Windows. macOS and Linux supported too.

MCP Tools & Claude-Native

Auto-registered MCP server and two installed skills (second-opinion, sidecar). Claude drives Amicus; you stay in your editor.

Context Sharing & Safety

Conversation history passed automatically. Conflict detection warns when files changed externally — a fold never silently overwrites newer work.

Session Persistence

Every session is saved. List, resume, or chain new investigations on previous findings. Full JSON output for scripting and agent consumption.

Before You Start

Prerequisites
& What It Costs.

A few things to have in place — and an honest look at what a run actually costs you.

Prerequisites

What you need

Node.js 18+ — run node --version to check.
An active Claude Code or Cowork session — Amicus is orchestrated by Claude; it is not a standalone chatbot.
At least one paid model API key — OpenRouter, or a direct Google / OpenAI / Anthropic / DeepSeek key.

What It Costs

~5–8 model calls per council

A sidecar is a single model call; a full council is typically ~5–8 paid model calls (3 reviewers across 2 fan-out waves + 1 chair). Amicus estimates the cost up front and a budget gate blocks runaway spend.

Want to try it for nothing? OpenRouter offers free model variants — route a council through those and you pay zero for tokens.

You pay providers for tokens. Amicus is free & open-source.

Foundation

Built on OpenCode.

Amicus builds on OpenCode, not around it. OpenCode does the heavy lifting.

OpenCode

opencode.ai

OpenCode is the open-source AI coding engine that powers every Amicus session. It handles the conversation, tool use, native agent types, and the web UI. Amicus adds context sharing from Claude Code, session history, the fold workflow, MCP support, and the Electron shell. We don't reinvent the wheel.

opencode.ai

Every model.One verdict.