T3 Code Dev Stack
IntermediateOne open-source app for Claude Code, Codex, OpenCode, and other coding agents, on desktop and mobile.
Published 27 September 2026
About T3 Code Dev Stack
T3 Code is an open-source control surface for coding agents, built by Ping Labs, the studio behind T3 Chat. It has no model or agent loop of its own: it's a frontend for the agent CLIs you already run, signing in with the subscriptions or API keys you already hold, with no token reselling in between. Several agents can work at once, each thread writing to its own branch.
In this stack agents and models are separate choices, and both allow more than one pick. Claude Code is the default agent, with Codex, OpenCode, Cursor's agent, and Grok Build alongside it. Each agent has a home model family, but the model layer isn't welded to the agent: OpenCode reaches Gemini, DeepSeek, and nearly any other provider, Cursor's agent brings its own model menu, and Codex can be pointed at other providers too.
Review happens turn by turn. A diff viewer shows what the agent changed at each step, and a built-in terminal runs linters and builds without switching windows. When a thread is done, one button opens the pull request on GitHub with a generated title, body, and changelog, including draft and stacked pull requests, which is why GitHub sits in the core rather than as a swappable option.
By default T3 Code and its agents run on your own computer, and the iOS and Android apps reach it from anywhere through T3 Connect, Ping Labs' relay, or with a pairing code over your home network or a Tailnet; that computer has to stay on while the agents work. For runs that outlast a closed laptop, the desktop app can connect to a VPS or home server over SSH, start T3 Code there, and list those remote threads next to your local ones, so the interface stays the same while the agents run elsewhere. T3 Code is free, MIT-licensed, and runs on macOS, Windows, and Linux.
Key Features
- ✓One interface for several agent CLIs, using the subscriptions and API keys you already hold
- ✓Pick one or more agents: Claude Code, Codex, OpenCode, Cursor's agent, and Grok Build
- ✓Pick one or more models, from Claude, OpenAI, and Grok to Gemini and DeepSeek through OpenCode
- ✓Each agent thread writes to its own branch, so parallel work stays separated
- ✓Turn-by-turn diff viewer showing what the agent changed at each step
- ✓One-button GitHub pull requests with generated title, body, and changelog, including draft and stacked PRs
- ✓Desktop apps for macOS, Windows, and Linux, plus iOS and Android apps
When to Use T3 Code Dev Stack
- →Replacing several agent terminals with one app while keeping the agents themselves unchanged
- →Comparing Claude Code, Codex, and Grok Build on the same task and shipping the best branch
- →Splitting a large agent-written change into stacked pull requests that are easier to review
- →Following agent progress from a phone while away from the desk
- →Developers who want an open-source alternative to closed agent orchestrators
Pros
- Free and MIT-licensed, with no proxy or markup between you and the agent providers
- Covers desktop and mobile on every major platform from one open codebase
- Turn-by-turn diffs make it easier to see where an agent went wrong, not just the final result
- Stacked and draft pull requests keep large agent changes reviewable
Cons
- An early-stage project that is still maturing, with rough edges reported by users
- Supports five named agents, fewer than the broadest orchestrators
- The Electron desktop app is heavier than native or terminal-based alternatives
- No hosted cloud, so running agents with the laptop closed means keeping a server of your own
Coding Agent Options for T3 Code Dev Stack
T3 Code Dev Stack with Claude Code
Claude Code is the default agent here: Anthropic's CLI, strong at multi-file changes and a natural pick for the hardest thread in a parallel run. It bills through a Claude plan from $20/month or per token via the API. It's built for Claude models; DeepSeek and GLM publish Anthropic-compatible endpoints it can point at, outside Anthropic's support.
T3 Code Dev Stack with Codex
OpenAI's agent CLI, included with every ChatGPT plan down to the free tier. A Codex thread next to a Claude Code thread on the same task is the easiest way to compare two vendors. It defaults to OpenAI's models but officially supports other providers through its config, including OpenRouter and local models.
T3 Code Dev Stack with OpenCode
Open-source and model-agnostic, OpenCode is what makes the model layer flexible: it runs Gemini, DeepSeek, Kimi, local models, or nearly any other provider with an API key, billed per token by that provider. The usual pick for cheap, high-volume threads next to a premium agent.
T3 Code Dev Stack with Cursor CLI
Cursor's agent driven from T3 Code, for developers already on a Cursor plan ($20/month and up). It brings Cursor's model menu, Grok by default with Claude, GPT, and Gemini available, into T3 Code without another subscription.
T3 Code Dev Stack with Grok Build
xAI's coding CLI, which splits a task across its own sub-agents in separate worktrees. Add it for a third vendor's take on a hard task; it's built around xAI's Grok models and bills through an xAI plan or API key.
These are highlighted picks. To see all the tools, check the AI Coding Agents category.
LLM Options for T3 Code Dev Stack
T3 Code Dev Stack powered by Claude
Claude is the default model, the one Claude Code is built for, and a common choice for the thread doing the most careful reasoning. OpenCode and Cursor's agent can use Claude models too, so several threads on different agents can share one model family.
T3 Code Dev Stack powered by OpenAI
OpenAI's GPT models are Codex's home family and are also reachable through OpenCode and Cursor's agent. Picking them alongside Claude gives a second vendor's approach to the same problem, and a ChatGPT plan covers Codex usage without per-token billing.
T3 Code Dev Stack powered by Grok
xAI's Grok models, the family Grok Build is built around and Cursor's default. Add Grok when a Grok Build or Cursor thread is in the mix, or reach it through OpenCode with an xAI API key for a fast option on routine tasks.
T3 Code Dev Stack powered by Google Gemini
Gemini's long context window suits repository-wide questions and large refactors, and its Flash models are fast and cheap for routine work. In this stack it's reached through OpenCode or Cursor's agent.
T3 Code Dev Stack powered by DeepSeek
An open-weight model priced far below the premium providers per token, the usual choice for bulk threads like test generation and mechanical refactors across many files. It runs through OpenCode, or behind Claude Code via DeepSeek's Anthropic-compatible endpoint.
These are highlighted picks. To see all the tools, check the LLM category.
T3 Code Dev Stack Add-ons
Each addition below extends this stack with a capability the base stack works fine without. None are required: include the ones your product actually needs when building this stack, and skip the rest.
Server Add-ons
Add a server when you want the agent to keep running somewhere other than your own laptop, reachable at any time and not tied to your machine staying on. Some agents also offer their own managed cloud sessions as an alternative to self-hosting; check the stack's own description for details.
T3 Code Dev Stack with Hetzner
Hetzner is the default here: a 4 GB server, Claude Code's minimum for one agent, is about €5.49/mo, and 8 GB or more suits several threads for little extra. The T3 Code desktop app can start its server on the machine over SSH, and data centers in Germany, Finland, the US, and Singapore let you pick one close to you.
T3 Code Dev Stack with DigitalOcean
DigitalOcean's 4 GB Droplet is $24/mo and 8 GB $48/mo, more per gigabyte than Hetzner, with more regions and one-click snapshots and backups. Its 1-Click Codex CLI, OpenCode, and Grok Build Droplets arrive with three of T3 Code's agents installed; the $4 plan's 512 MB can't run one.
T3 Code Dev Stack with Hostinger
Hostinger's KVM 1 has 4 GB of RAM at $6.49/mo introductory, renewing at $11.99/mo, enough for one agent, and KVM 2 doubles the memory for parallel threads. Its Claude Code VPS template preinstalls T3 Code's default agent, so the desktop app only has to start T3 Code's server there.
These are highlighted picks. To see all the tools, check the Hosting & Cloud category.
Remote Access Add-ons
Add remote access when you want to reach an agent running on another machine — a VPS or a home server — without exposing it to anyone but you.
T3 Code Dev Stack with Tailscale
Add Tailscale as the private alternative to T3 Connect: T3 Code pairs directly with a machine on your Tailnet, so a server stays reachable from your laptop and phone without any relay or public exposure. The personal tier is free.
Code Review Add-ons
Add code review when you want an AI reading every pull request before it merges: it comments on the diff so bugs, security issues, and inconsistencies surface before a human has to catch them.
T3 Code Dev Stack with CodeRabbit
Add CodeRabbit when several agents are opening pull requests faster than one person can read them: it reviews every GitHub or GitLab diff with a summary and line-level comments, with a free tier for open-source repos.
T3 Code Dev Stack with Greptile
Add Greptile when parallel agents produce more diffs than one person can review: it indexes the whole repository so its comments carry context beyond the changed files, and it can be self-hosted for full control.
T3 Code Dev Stack with Cursor Bugbot
Add Cursor Bugbot when a bug-focused reviewer is enough: it reviews GitHub or GitLab pull requests whatever wrote the code, though it bills through a Cursor account and plan rather than its own subscription.
CI/CD Add-ons
Add CI/CD when you want a dedicated pipeline for running tests, linting, or multi-stage builds before a deploy goes out. Many hosting platforms already redeploy automatically on every push on their own — a CI/CD tool adds the most value on top of that by gating the deploy on a passing test suite, and matters even more when the hosting choice does not auto-deploy at all, such as a self-hosted server.
T3 Code Dev Stack with GitHub Actions
Add GitHub Actions when several agents are opening pull requests at once and every branch should be tested automatically before you spend time reviewing it.
These are highlighted picks. To see all the tools, check the CI/CD Pipelines category.
Containerization Add-ons
Add containerization when you want the app packaged the same way across local development, staging, and production, or need to deploy somewhere that isn't a managed serverless platform.
T3 Code Dev Stack with Docker
Add Docker when each agent's workspace needs its own database or dev server: a Compose file per workspace keeps ports and data from colliding when several agents run the app at once in T3 Code.
These are highlighted picks. To see all the tools, check the Containerization category.
Frequently Asked Questions about T3 Code Dev Stack
Can agents keep working when my laptop is closed?
Only if T3 Code runs on another machine. The desktop app can start T3 Code on any server it reaches over SSH and shows those threads like local ones, while the agents keep running on the server with your laptop closed. The iOS and Android apps reach the same server through T3 Connect from any network, or by pairing code over your own network or Tailnet. T3 Connect is only the connection: it doesn't run agents, and T3 Code has no hosted cloud. Without a server, the phone apps steer threads on your desktop, which has to stay on.
Which agents can run which models?
Most combinations work, with varying effort. OpenCode takes almost any provider's key, and Cursor's agent offers a menu of Grok, Claude, GPT, and Gemini models on a Cursor plan. Codex defaults to OpenAI but officially supports other providers through its configuration, local models included. Claude Code is built for Claude, and Anthropic supports only Claude models, though DeepSeek and GLM publish Anthropic-compatible endpoints some developers point it at. Grok Build is built around Grok.
What are stacked pull requests, and why use them with agents?
A stacked pull request splits one large change into a chain of smaller PRs, each built on the one before, reviewed and merged in order. Agents tend to produce large diffs in one go, and a reviewer can check five focused PRs more carefully than one sprawling one. T3 Code can open the chain from a finished thread.
Is T3 Code stable enough for daily work?
It's an early project that changes quickly, so expect rough edges between releases. The risk is low, though: T3 Code only drives agent CLIs you already have, so if something breaks you can keep working with the same agent in a terminal and come back later.
How is T3 Code different from Conductor?
Conductor is a closed-source Mac app with four agents and a paid hosted cloud that keeps agents running with your Mac off. T3 Code is free and MIT-licensed, runs on macOS, Windows, and Linux with mobile apps, adds Grok Build and turn-by-turn diffs with stacked pull requests, and keeps agents running away from your desk on a server of your own rather than a hosted cloud.
Stacks Related to T3 Code Dev Stack
Paseo Dev Stack
Developer
Run Claude Code, Codex, and other coding agents on your own machine with Paseo and steer them from any device.
Claude Code Dev Stack
DeveloperClaude Code in the terminal, desktop app, or web, paired with Claude and the add-ons to run it anywhere.
Conductor Dev Stack
Developer
Run Claude Code, Codex, and Cursor's agent in parallel on your Mac with Conductor, each in its own workspace.
Orca Dev Stack
Developer
Run Claude Code, Codex, and other coding agents side by side in Orca, each on its own branch and model.
Scores
Popularity3/5
T3 Code benefits from its studio's large developer audience, but agent orchestrators remain a young category used mainly by AI-heavy developers.
Learning Curve2/5
A familiar chat-and-diff GUI hides worktrees and branch management, and pull requests open with one button, so the main learning is the agents themselves and when to run them in parallel.
Flexibility4/5
Six agent CLIs, a model layer opened up by OpenCode, and a forkable MIT codebase give plenty of room, though the pull-request flow is built for GitHub.
Performance3/5
Parallel agent threads save real time on independent tasks, but the Electron desktop app is heavier than native or terminal-based orchestrators.
Portability5/5
MIT-licensed and forkable, with clients for macOS, Windows, Linux, iOS, and Android, and no lock-in to any single agent or model provider.
Tools in the T3 Code Dev Stack Stack
Add-ons (optional — add any, or none)
Server
Remote Access
Code Review
CI/CD
Containerization
T3 Code Dev Stack Pricing
T3 Code itself is free and open source, so the bill comes from the agents. Claude Code starts at $20/month with no free tier, Codex has a limited free tier with Plus at $20/month, both reach $200/month, Cursor is $20/month, and OpenCode is free but bills per token through the provider you connect. Several agents usually means several subscriptions. A server for headless runs is about €5.49/month for 4 GB and more for several agents; GitHub and Tailscale's personal tier are free.
MIT-licensed desktop and mobile apps with no paid tiers.
Claude Code from $20/mo (no free tier); Codex free tier, Plus $20/mo; both up to $200/mo; Cursor $20/mo. One subscription per agent.
Model-agnostic agents bill per token with the provider you connect; open-weight models like DeepSeek and Kimi cost a fraction of premium ones.
Free for individual developers; GitHub Team is $4/mo per user for more seats and controls.
GitHub Actions includes free monthly minutes before paid usage kicks in.
Docker Desktop is free for personal use; Docker Pro adds team features from $5/mo.
Tailscale's personal tier is free for up to 100 devices.
4 GB runs one agent, 8 GB or more suits several: Hetzner from €5.49, Hostinger $6.49 intro ($11.99 renewal), DigitalOcean $24–48.