The work lives in a chat window
Close the tab and the plan, the decisions and half the context are gone. Nothing is queryable afterwards.
Your coding agents,
on your own server.
Agentic Control runs Claude Code, Codex and OpenCode as one team on a VPS you own. One agent plans and reviews, another implements, and you approve the pull request — from your laptop or your phone.
Works with the subscriptions you already pay for: Claude, ChatGPT, OpenCode, or any model through OpenRouter. No hosted account in between — the workspace, the sessions and the history stay on your server.
Why
Agents are capable. The scaffolding around them is not.
The bottleneck is no longer model quality. It is state, handoff and trust.
Every agent starts from zero
A fresh clone, a fresh environment, a fresh misunderstanding of the codebase — repeated on every task.
Two agents, no handoff
The model that plans well is rarely the one that edits best, and which is which changes every month. Moving work between them by hand is slow and lossy.
Your code sits on someone else's disk
Hosted agent platforms need your repository, your keys and your database. For a lot of work that is simply not an option.
How it works
One workspace. A team of agents. Every step on the record.
You pick the team per project — Claude Code orchestrating Codex, or Codex orchestrating OpenCode. The platform keeps the state; the agents do what they are already good at.
- 01
Write the task in a chat
Connect a GitHub repository — the GitHub App is created from the panel in a few clicks — and describe the change the way you would to a colleague.
- 02
Plan and delegate
The orchestrator reads the workspace, writes the plan and hands the implementation to an executor. The handoff is an event, and it is recorded.
- 03
Implement in the workspace
The executor edits one persistent project workspace — the same checkout, dependencies and tooling as last time — and commits as the app's bot, not as you.
- 04
Review and revise
The orchestrator checks the diff against the plan and sends it back with notes until it holds. You see every revision and every file it touched.
- 05
Approve and publish
Approve the review and the panel opens the pull request. An empty repository gets its first commit on main. Merge it where you always do.
Who it is for
People who already pay for agents and want them to work while they don't.
For a developer on their own
- Your Claude and ChatGPT subscriptions do the work. The panel shows how much of each limit is left before you start a run, not after.
- Start a task from your phone on the way, review it from your laptop when you sit down.
- New agent versions are installed beside the current one, checked on your server and offered with one button.
For a small team
- One server you control: the code, the sessions and the history stay on your infrastructure.
- Every agent runs as its own system user, with no access to the database or to another agent's login.
- Commits come from a bot identity, pull requests go through your own GitHub App, and every release is signed.
One person operates the panel today; seats for the rest of the team are next.
Architecture
One VPS. One process facing the internet.
Nothing is managed for you, because nothing needs to be. The installation is a Node build, a database on a Unix socket and a reverse proxy.
Request path
Browser or phone
Caddy
Next.js standalone
PostgreSQL 17
No password anywhere in the path
The panel and each worker connect over a Unix socket. The operating-system user and the PostgreSQL role are mapped one to one, so there is no connection string to leak and no secret to rotate.
Agents cannot reach the database
Each runtime executes as its own system account, sandboxed, with no database role and no way to read another runtime's login. Agents edit the workspace; the control plane records what happened.
Releases are signed and verifiable
An installation is a single artifact with a manifest and a signature you can check before it ever runs on your machine.
Status
Where the product actually is.
Agentic Control is in private beta: it runs every day on one production server — its author's. This is the honest state of it.
Panel and workflow
Projects, chats, orchestrator and executor teams, review, revisions and publishing to GitHub. Used every day by its author.
Agent runtimes
Claude Code, Codex and OpenCode. A new version is qualified on your server before you promote it, and the model list refreshes itself.
Installation
One command on a clean Ubuntu 24.04 VPS (x86-64, 2 GB of memory, 10 GB of disk): database, services, TLS and the panel.
Signed releases and updates
Every release is signed. An update checks the signature, takes a backup first and keeps the previous version for a rollback.
Several users per server
One operator per installation today.
General availability
Access is granted case by case while the product proves itself on servers other than its author's.
Early access
Run it on your own server.
Private beta, granted case by case. Tell us what you would run it on — you will hear back from a person, not an autoresponder.