starkomand.com

mx

eforward2.registrar-servers.com

eforward1.registrar-servers.com

eforward3.registrar-servers.com

eforward4.registrar-servers.com

eforward5.registrar-servers.com

ns

coraline.ns.cloudflare.com

chance.ns.cloudflare.com

caa

txt

google-site-verification=6w0A0LQeOz0lvDhplo1a94K-pK1KfQ-U82Kgf6BAwhg

v=spf1 include:spf.efwd.registrar-servers.com ~all

dmarc

v=DMARC1; p=reject;

# Public product information is available to search and retrieval crawlers.
# Cloudflare may prepend its managed rules; training preferences remain there.
User-agent: *
Allow: /
Disallow: /api/
Disallow: /checkout
# Cloudflare's own endpoints. Its email obfuscation rewrites each mailto link to
# /cdn-cgi/l/email-protection#…, and a crawler that drops the fragment gets a 404.
Disallow: /cdn-cgi/

Sitemap: https://starkomand.com/sitemap.xml

# Starkomand

> Starkomand is an agentic development environment (ADE) and AI agent orchestration app for macOS on Apple silicon. It carries coding work from an issue through implementation and review, with agents, saved task briefs, handoff history and review evidence on one canvas.

Developers can run Claude Code, Codex, Cursor, Gemini CLI, Grok CLI, GitHub Copilot, OpenCode (including GLM via Z.ai), Amp, Pi, Aider, Goose and any other configured CLI agent side by side using their existing provider sign-ins. Starkomand does not resell tokens or ask for API keys. Agent subscriptions and usage are billed separately by the providers.

One agent is enough to start. Optional model handoffs pass a focused brief to a fresh agent, and Auto model routing picks each step's model and effort across the user's Claude and Codex subscriptions, saying why. Bounded review loops retain findings and checks against a repository revision, pause when files change, and leave task acceptance to the engineer.

Each workspace keeps its own agents, project folder, layout and backdrop. Tapping Space flips every window to show where its agent stands, what it said last, and the files, dev servers and pull requests it left. Hiding the canvas keeps agents working while the local server runs, and with Settings → Lid closed on, a MacBook stays awake with its lid closed while agents work. Restarting the Mac ends live sessions but preserves the canvas, notes and task history.

Voice control and hand gestures are optional. Local dictation requires an optional download. Workspace data, local dictation and hand tracking stay on the Mac by default; agents and connected services use their providers. Window backs and Auto routing call no model. A work summary sends terminal context to the provider selected in Privacy settings, only when the user asks for one. Optional cloud dictation sends audio to OpenAI. See the privacy policy for trial verification and other connected-service flows.

The 3-day trial includes every feature. Continuing after the trial requires Pro: a one-time purchase for one Mac, including a year of updates. The purchased version remains usable after that year. Launch prices depend on remaining seats, so consult the current pricing section instead of assuming a fixed price. Use the homepage download link for the current release.

## Product

- [Product guide in Markdown](https://starkomand.com/index.md): Start here for the product overview, supported platforms, setup, workflows, privacy and licensing in plain text.
- [The whole site in one file](https://starkomand.com/llms-full.txt): The product guide, comparisons, blog posts, agent pages and policies as Markdown, for reading everything at once.
- [Starkomand homepage](https://starkomand.com/): An interactive, scripted canvas demo, the product overview and the current Mac download.
- [How it works](https://starkomand.com/#workflow): One scene per move from the homepage demo, captured from the app: deliver an issue to an agent, draw a diagram, hand off with Auto picking each model, talk to the crew by voice, flip every window to catch up.
- [Canvas features](https://starkomand.com/#features): Lid-closed work, the Ports widget, browser previews, workspaces, the file guard, agent messages, the Flight board, usage limits, persistent agents, hand gestures and backdrops.
- [Supported agents](https://starkomand.com/#agents): Every agent CLI the canvas launches, with its command.
- [Voice control](https://starkomand.com/#scene-voice): Optional spoken commands for directing agents by name.
- [Local workspace](https://starkomand.com/#local): What stays on the Mac and which features use connected services.
- [Pricing and trial](https://starkomand.com/#pricing): Current Pro offer, launch pricing and trial details.
- [What is an ADE?](https://starkomand.com/ade/): What an agentic development environment is, how it differs from an IDE, and where a canvas-based Mac ADE fits.
- [AI agent orchestration for coding](https://starkomand.com/agent-orchestration/): Roles, briefs, bounded review loops and file ownership when several coding agents work on one task.
- [Compared with Conductor, Superset, Emdash, Maestri, AgentGrid, CNVS, 49Agents, Zeron and hosted agents](https://starkomand.com/compare/): Feature-by-feature table with prices, where the code runs and which agents each tool supports, dated from the products' public pages.
- [Starkomand vs AgentGrid](https://starkomand.com/compare/agentgrid/): The two products alone, the rows where they differ and which to pick.
- [Starkomand vs Maestri](https://starkomand.com/compare/maestri/): The two products alone, the rows where they differ and which to pick.
- [Starkomand vs Conductor](https://starkomand.com/compare/conductor/): The two products alone, the rows where they differ and which to pick.
- [Starkomand vs Superset](https://starkomand.com/compare/superset/): The two products alone, the rows where they differ and which to pick.
- [Starkomand vs Emdash](https://starkomand.com/compare/emdash/): The two products alone, the rows where they differ and which to pick.
- [Starkomand vs CNVS](https://starkomand.com/compare/cnvs/): The two products alone, the rows where they differ and which to pick.
- [Starkomand vs 49Agents](https://starkomand.com/compare/49agents/): The two products alone, the rows where they differ and which to pick.
- [Starkomand vs Zeron](https://starkomand.com/compare/zeron/): The two products alone, the rows where they differ and which to pick.
- [AgentGrid alternatives](https://starkomand.com/compare/agentgrid-alternatives/): The other apps for running coding agents instead of AgentGrid, each with its price, platform and who it suits.
- [Maestri alternatives](https://starkomand.com/compare/maestri-alternatives/): The other apps for running coding agents instead of Maestri, each with its price, platform and who it suits.
- [Conductor alternatives](https://starkomand.com/compare/conductor-alternatives/): The other apps for running coding agents instead of Conductor, each with its price, platform and who it suits.
- [Superset alternatives](https://starkomand.com/compare/superset-alternatives/): The other apps for running coding agents instead of Superset, each with its price, platform and who it suits.
- [Emdash alternatives](https://starkomand.com/compare/emdash-alternatives/): The other apps for running coding agents instead of Emdash, each with its price, platform and who it suits.
- [CNVS alternatives](https://starkomand.com/compare/cnvs-alternatives/): The other apps for running coding agents instead of CNVS, each with its price, platform and who it suits.
- [49Agents alternatives](https://starkomand.com/compare/49agents-alternatives/): The other apps for running coding agents instead of 49Agents, each with its price, platform and who it suits.
- [Zeron alternatives](https://starkomand.com/compare/zeron-alternatives/): The other apps for running coding agents instead of Zeron, each with its price, platform and who it suits.
- [Run multiple Claude Code agents in parallel](https://starkomand.com/compare/run-multiple-claude-code-agents/): The git worktree setup for running several Claude Code sessions at once, then every app above with how it runs Claude Code, its price and platform.
- [Codex vs Claude Code](https://starkomand.com/compare/codex-vs-claude-code/): The two terminal agents side by side on benchmarks, plan and API prices, usage limits and permission defaults, dated and sourced, with how to run both.
- [Frequently asked questions](https://starkomand.com/#faq): Supported agents, resource use, session persistence and licensing.

## Agents

- [Multi-agent orchestration for coding CLIs](https://starkomand.com/agents/): Run 12 coding agent CLIs side by side on one Mac canvas. Compare their launch commands, canvas tools, usage rings and sign-in recovery.
- [Claude Code GUI for Mac: run multiple sessions](https://starkomand.com/agents/claude-code/): A Claude Code GUI and desktop app for Mac: run multiple Claude Code sessions side by side in real terminals, signed in as you, with every tool step in view.
- [Codex CLI GUI for Mac: run multiple sessions](https://starkomand.com/agents/codex/): A Codex CLI GUI for Mac: run multiple OpenAI Codex sessions in real terminals with your ChatGPT sign-in, plus diagrams, previews and handoffs to Claude Code.
- [Cursor CLI on Mac: run agents side by side](https://starkomand.com/agents/cursor/): Run cursor-agent in real terminals on one Mac canvas, on your own Cursor plan. Usage ring, diagrams and previews, no editor juggling.
- [Gemini CLI GUI for Mac: run multiple sessions](https://starkomand.com/agents/gemini-cli/): Run several Google Gemini CLI sessions on one Mac canvas, each in a real terminal with your Google sign-in, next to Claude Code, Codex and the rest.
- [Run multiple Grok CLI sessions on a Mac](https://starkomand.com/agents/grok/): Run xAI’s Grok CLI in real terminals on one Mac canvas, several sessions at once, with diagrams, browser previews and handoffs to your other agents.
- [GLM Coding Plan on Mac: run it via OpenCode](https://starkomand.com/agents/glm/): GLM has no CLI, so run the Z.ai coding plan through OpenCode on one Mac canvas, next to Claude Code and Codex, on your own Z.ai key.
- [Run multiple GitHub Copilot CLI sessions](https://starkomand.com/agents/copilot/): Run several GitHub Copilot CLI sessions on one Mac canvas, each in a real terminal with your GitHub sign-in, a Copilot usage ring and handoffs.
- [OpenCode GUI for Mac: run multiple sessions](https://starkomand.com/agents/opencode/): Run several OpenCode sessions on one Mac canvas, each a real terminal on whichever provider you connected, with diagrams, previews and a Go usage ring.
- [Run multiple Amp coding agent sessions](https://starkomand.com/agents/amp/): Run several Amp CLI sessions on one Mac canvas, each in a real terminal with your Amp sign-in, next to Claude Code, Codex and your other coding agents.
- [Run the Pi coding agent on a Mac canvas](https://starkomand.com/agents/pi/): Run several Pi coding agent sessions on one Mac canvas in real terminals. Pi has no MCP by design, so canvas tools arrive as a shell command it can run.
- [Aider GUI for Mac: run multiple sessions on one canvas](https://starkomand.com/agents/aider/): Run several Aider sessions on one Mac canvas with your own model keys. See diffs in real terminals, with diagrams and previews beside them.
- [Run multiple Goose agent sessions on a Mac](https://starkomand.com/agents/goose/): Run several Goose sessions on one Mac canvas in real terminals, with the provider you set up, plus diagrams, browser previews and handoffs to other agents.

## Integrations

- [Integrations for AI coding agents on Mac](https://starkomand.com/integrations/): Starkomand integrations: GitHub issues and pull requests, Excalidraw diagrams, persistent agents with Herdr, and 12 coding agent CLIs on one Mac canvas.
- [GitHub issues to Claude Code and Codex agents](https://starkomand.com/integrations/github/): Drop a GitHub issue on a Claude Code or Codex window to assign it, and keep your pull requests, CI and reviews on the same Mac canvas as the agents.
- [Excalidraw diagrams from Claude Code and Codex](https://starkomand.com/integrations/excalidraw/): Claude Code, Codex and other coding agents draw editable Excalidraw diagrams on your Mac canvas: architecture, flows and walkthroughs beside their windows.
- [Persistent Claude Code sessions with Herdr](https://starkomand.com/integrations/herdr/): Run Claude Code, Codex or any coding agent inside Herdr from the Starkomand canvas: it keeps working after you quit the app and comes back as a window.

## Use cases

- [Use cases for AI coding agents](https://starkomand.com/for/): How freelancers, agencies, enterprise developers, indie hackers, founders and maintainers run Claude Code, Codex and other coding agents on one Mac canvas.
- [AI coding agents for freelance developers](https://starkomand.com/for/freelance-developers/): Run Claude Code, Codex and other coding agents across client projects on your Mac. One workspace per client, with its own folder, agents and notes.
- [AI coding agents for software agencies](https://starkomand.com/for/agencies/): Give each developer a crew of coding agents across client projects: a workspace per client, GitHub issues dropped on agents, and a review before sign-off.
- [AI coding agents for enterprise developers](https://starkomand.com/for/enterprise-developers/): Run Claude Code, Codex or Copilot CLI on your company codebase from one Mac canvas: map large systems, work tickets in parallel, review before the PR.
- [AI coding agents for indie hackers](https://starkomand.com/for/indie-hackers/): Ship side projects with a crew of coding agents on your Mac: one plans, one builds, one reviews, and the preview opens beside them. Pay once.
- [AI coding agents for SaaS founders](https://starkomand.com/for/saas-founders/): Run your SaaS codebase with coding agents on your Mac: customer bugs dropped on agents, a second agent reviewing each fix, and the PRs that need you.
- [AI coding agents for open source maintainers](https://starkomand.com/for/open-source-maintainers/): Work through your GitHub issues with coding agents: drop an issue on Claude Code or Codex, have a second agent review the fix, and see which PRs need you.

## Tools

- [Model routing planner](https://starkomand.com/tools/model-routing-planner/): Free tool: pick your Claude, ChatGPT, Google AI, X (Grok and Cursor), GitHub Copilot, Kimi and GLM subscriptions and a job, and see which model (Fable 5.1, Opus 5.5, GPT-6 Astra, GPT-6.1 Sol, Gemini 4 Argon, Grok 4.7, Kimi K3, GLM-5.3) should plan, build, review and test it, with the CLI command for each step.

## Policies and support

- [Privacy policy](https://starkomand.com/privacy/): Data handling, connected providers and optional services.
- [Terms and conditions](https://starkomand.com/terms/): License terms, updates and refund policy.
- [Cookie policy](https://starkomand.com/cookies/): The site sets no cookies of its own; which security cookies its host can set.
- [Find your license key](https://starkomand.com/license/): Recovery instructions for existing customers.

## Blog

- [What to do when Claude Code hits its usage limit](https://starkomand.com/blog/claude-code-usage-limit-continue-in-codex/): Claude Code usage limit reached mid-task? See your session and weekly runway, then continue the same work in Codex from the recorded conversation.
- [How to run Claude Code and Codex side by side on a Mac](https://starkomand.com/blog/run-claude-code-and-codex-side-by-side/): Run Claude Code and Codex side by side on a Mac without edit collisions: split the roles, isolate writers in Git worktrees, and see who finished.
- [Claude Code voice mode: voice coding across agents](https://starkomand.com/blog/superwhisper-alternative-for-coding-agents/): Learn Claude Code voice mode, then try voice coding across named agents in Starkomand with on-device Parakeet or OpenAI dictation. Hold a key to speak.
- [Hand off a task from Claude Code to Codex without losing the plan](https://starkomand.com/blog/handoff-csv-export-claude-code-to-codex/): How to hand off a task from Claude Code to Codex: plan a CSV export, then pass the objective, acceptance criteria and brief to Codex in Starkomand.
- [How to keep track of multiple AI coding agent sessions](https://starkomand.com/blog/resume-two-projects-with-agent-recaps/): Back from a break with agents still running? Keep track of multiple AI coding agent sessions with window states, spoken notices and short recaps.
- [Have a second AI agent review code changes: a search race fix](https://starkomand.com/blog/review-search-race-condition-with-second-agent/): Have a second AI agent review code changes: hand a Claude Code fix to a Codex reviewer with a factual brief, then make the final call yourself.
- [Why I put AI coding agents on an infinite canvas](https://starkomand.com/blog/why-i-put-coding-agents-on-an-infinite-canvas/): An infinite canvas for AI coding agents keeps the issue, diagram, preview and reviewer beside the terminal. Why I built Starkomand that way.