npx maxsimcli@latest

Getting Started

npx maxsimcli@latest

The installer prompts you to choose:

1. Runtime — Claude Code, OpenCode, Gemini, Codex, or all
2. Location — Global (all projects) or local (current project only)

Verify with:

• Claude Code / Gemini: /maxsim:help
• OpenCode: /maxsim-help
• Codex: $maxsim-help

npx maxsimcli --claude --global    # Claude Code → ~/.claude/
npx maxsimcli --opencode --global  # OpenCode → ~/.config/opencode/
npx maxsimcli --gemini --global    # Gemini CLI → ~/.gemini/
npx maxsimcli --codex --global     # Codex → ~/.codex/
npx maxsimcli --all --global       # All runtimes

Live Dashboard

MAXSIM ships with a real-time web dashboard — bundled inside the CLI, no separate setup needed.

npx maxsimcli dashboard

The dashboard opens in your browser and updates instantly as .planning/ files change via WebSocket.

Dashboard features:

• Phase overview — progress bars, milestone stats, and completion percentage
• Phase drill-down — expand any phase to see individual plan tasks with toggleable checkboxes
• Inline Markdown editor — edit plan files directly in the browser (CodeMirror, Ctrl+S to save)
• Todos panel — create, complete, and delete todos
• Blockers panel — view and resolve blockers from STATE.md
• STATE.md editor — edit project state inline
• Auto-launches during /maxsim:execute-phase so you always have a live view
• Idempotent — running the command again when a server is already up does nothing
• LAN sharing — share your dashboard with teammates on the same network

Network / LAN Sharing

npx maxsimcli dashboard --network

Enables LAN sharing so anyone on your local network (or Tailscale VPN) can open the dashboard in their browser. MAXSIM:

• Detects your local IP and Tailscale IP automatically
• Configures firewall rules on Windows (netsh) and Linux (ufw/iptables) with one command
• Generates a QR code so you can open the dashboard on your phone in seconds

How It Works

1. Initialize Project

/maxsim:new-project

Questions → Research → Requirements → Roadmap. One command captures your idea, scopes v1/v2, and creates a phased build plan.

2. Discuss Phase

/maxsim:discuss-phase 1

Shape the implementation before anything gets built. The system identifies gray areas and asks until your vision is clear.

3. Plan Phase

/maxsim:plan-phase 1

Researches how to implement, creates atomic task plans, and verifies them against requirements.

4. Execute Phase

/maxsim:execute-phase 1

Runs plans in parallel waves with fresh context per plan. Each task gets its own atomic commit. Verifies the codebase delivers what the phase promised.

5. Verify Work

/maxsim:verify-work 1

Walk through testable deliverables one by one. If something's broken, the system creates fix plans automatically.

6. Repeat → Ship

/maxsim:complete-milestone
/maxsim:new-milestone

Loop discuss → plan → execute → verify until done. Archive the milestone and start the next version.

Commands

Core Workflow

Navigation & Utilities

Phase Management

Milestone & Quality

Todos

Live Dashboard

Settings

Configuration

Project settings live in .planning/config.json, created during /maxsim:new-project or editable via /maxsim:settings.

{
  "model_profile": "balanced",
  "branching_strategy": "none",
  "commit_docs": true,
  "research": true,
  "plan_checker": true,
  "verifier": true,
  "parallelization": true,
  "brave_search": false
}

Model Profiles

MAXSIM has 4 model profiles that control which Claude model each of the 11 specialized agents uses:

tokenburner assigns Opus to every single agent. Use it when cost is no concern and you want maximum quality end-to-end.

Switch profiles at any time:

/maxsim:set-profile quality
/maxsim:set-profile balanced
/maxsim:set-profile budget
/maxsim:set-profile tokenburner

You can also override individual agents in config.json:

{
  "model_profile": "balanced",
  "model_overrides": {
    "maxsim-planner": "opus",
    "maxsim-executor": "opus"
  }
}

Hook System

MAXSIM installs three compiled hooks into Claude Code:

The context bar in the statusline shows a 10-segment progress indicator that turns red and blinks when context is above 95% — giving you a clear signal to spawn a new session before quality degrades.

Agents

11 specialized subagents, each with fresh context and a single responsibility:

Acknowledgments

Inspired by GSD (Get Shit Done).

License

MIT License. See LICENSE for details.