Multica

AI-native task management platform — like Linear, but with AI agents as first-class citizens.

Prerequisites

• Node.js (v20+)
• pnpm (v10.28+)
• Go (v1.26+)
• Docker

Quick Start

# 1. Install dependencies
pnpm install

# 2. Copy environment variables for the shared main environment
cp .env.example .env

# 3. One-time setup: start DB and run migrations
make setup

# 4. Optional: load example data
make seed

# 5. Start backend + frontend
make start

Open your configured FRONTEND_ORIGIN in the browser. By default that is http://localhost:3000.

Default behavior now prefers the shared main environment in .env. If you want an isolated environment for a Git worktree, generate .env.worktree and use the explicit worktree targets:

make worktree-env
make setup-worktree
make start-worktree

This lets you keep .env connected to your main database while using .env.worktree only for isolated feature testing.

Project Structure

├── server/             # Go backend (Chi + sqlc + gorilla/websocket)
│   ├── cmd/            # server, daemon, migrate, seed
│   ├── internal/       # Core business logic
│   ├── migrations/     # SQL migrations
│   └── sqlc.yaml       # sqlc config
├── apps/
│   └── web/            # Next.js 16 frontend
├── packages/           # Shared TypeScript packages
│   ├── ui/             # Component library (shadcn/ui + Radix)
│   ├── types/          # Shared type definitions
│   ├── sdk/            # API client SDK
│   ├── store/          # State management
│   ├── hooks/          # Shared React hooks
│   └── utils/          # Utility functions
├── Makefile            # Backend commands
├── docker-compose.yml  # PostgreSQL + pgvector
└── .env.example        # Environment variable template

Commands

Frontend

Command	Description
pnpm dev:web	Start Next.js dev server (uses FRONTEND_PORT, default 3000)
pnpm build	Build all TypeScript packages
pnpm typecheck	Run TypeScript type checking
pnpm test	Run TypeScript tests

Backend

Command	Description
make dev	Run Go server (uses PORT, default 8080)
make daemon	Run local agent daemon
make test	Run Go tests
make build	Build server & daemon binaries
make sqlc	Regenerate sqlc code from SQL

Database

Command	Description
docker compose up -d	Start PostgreSQL
docker compose down	Stop PostgreSQL
make migrate-up	Run database migrations
make migrate-down	Rollback database migrations
make seed	Seed example data
make worktree-env	Generate an isolated .env.worktree for the current worktree
make setup-main / make start-main	Force use of the shared main .env
make setup-worktree / make start-worktree	Force use of isolated .env.worktree

Environment Variables

See .env.example for all available variables:

• DATABASE_URL — PostgreSQL connection string
• COMPOSE_PROJECT_NAME — Docker Compose project name
• POSTGRES_DB / POSTGRES_PORT — Per-worktree PostgreSQL database and host port
• PORT — Backend server port (default: 8080)
• FRONTEND_PORT / FRONTEND_ORIGIN — Frontend port and browser origin
• JWT_SECRET — JWT signing secret
• MULTICA_APP_URL — Browser origin used when generating local runtime pairing links
• MULTICA_DAEMON_CONFIG — Optional path for the daemon's persisted local config
• MULTICA_WORKSPACE_ID — Optional dev override for the workspace id; normal usage should rely on browser pairing instead
• MULTICA_DAEMON_ID / MULTICA_DAEMON_DEVICE_NAME — Stable daemon identity for local runtime registration
• MULTICA_CODEX_PATH / MULTICA_CODEX_MODEL — Codex executable and optional model override for local task execution
• MULTICA_CODEX_WORKDIR — Default working directory used by the local Codex runtime
• GOOGLE_CLIENT_ID / GOOGLE_CLIENT_SECRET — Google OAuth (optional)
• NEXT_PUBLIC_API_URL — Frontend → backend API URL
• NEXT_PUBLIC_WS_URL — Frontend → backend WebSocket URL

Local Codex Daemon

The local daemon currently supports one local runtime type: codex.

1. Start the daemon with make daemon.
2. If the daemon does not already know its workspace, it prints a pairing link in the terminal.
3. Open that link in the browser, sign in, and choose the workspace that should own the local Codex runtime.
4. The daemon stores the approved workspace locally in MULTICA_DAEMON_CONFIG or ~/.multica/daemon.json.
5. The daemon registers the local Codex runtime via /api/daemon/register.
6. Create an agent in Multica and bind it to that runtime.
7. Assign an issue to the agent and move the issue to todo.
8. The daemon claims the task, runs codex exec, and reports the final comment back to the issue.

For local development you can still set MULTICA_WORKSPACE_ID directly to skip pairing, but that should be treated as a debug shortcut rather than the normal flow.