# AITJE Coder: Your own coding agents. With no external token costs.

Official page: https://aitje.com/en/products/aitje-coder

> AITJE Coder AI Station: coding agents, Git review and scheduled tasks on hardware or a server. From €799 including basic installation, excluding VAT and infrastructure.

Status: Available

AITJE Coder is a proprietary alternative to services such as Codex and Claude Code: a complete local coding environment that reads and modifies projects, executes commands and completes programming tasks iteratively.

AITJE has already done the legwork. Selected local coding models, an agent with review and pull requests, a web app for management and integration with tools such as Claude Code, OpenCode, Codex and Aider are all included, ready to use, in a single product.

## Who it is for

- Development agencies that work with coding agents on a daily basis
- Companies with their own development team
- Freelance developers and technical creators
- Professionals who wish to use their own or spare hardware
- Teams seeking greater control over models, infrastructure and costs

## Capabilities

### Terminal and CLI

Work from your own terminal with models running locally.

### AITJE-interface

A graphical interface for those who prefer not to work in the terminal.

### Supported harnesses

Connect Coder to Claude Code, OpenCode, Codex, Aider and other coding agents.

### Selected models

Light, fast or powerful: you choose according to the task, without needing to know the model names.

### Iterative working

Agents read and modify projects, execute commands and complete tasks.

### Works without an internet connection

The local core remains usable without a connection to an external model provider.

## Agent & review

From new task to assessed result

### Tasks and multiple repositories

Select the repository, start branch, model and web search. Where necessary, work on linked repositories within your own secure working folder.

### Live progress and stopping

Track files, commands, tests and errors. Stop a task with confirmation; a stalled loop will be given a limited number of restarts.

### Review by file

View changes side by side or one below the other. Provide an assessment and a comment for each file. Compare runs or have a change re-run.

### Approval and Git

A reviewer opens the pull request on GitHub, GitLab or Gitea. GitHub login using a code and a fork where write access is lacking are supported.

### Queue management, scheduling and rounds

Queue or pause the process. Schedule it once, daily, weekly or via cron. Start on a push or a GitHub/Linear issue and run multiple rounds.

### Priority for interactive work

If you are working in an editor or chat yourself, this takes precedence over batch tasks. The queue shows what is pending, running or awaiting review.

## Tools & models

The environment for your existing development work

### Terminal and editor

Configurations for Claude Code, OpenCode, Codex CLI, Aider, Cline, Roo Code, Kilo Code, Continue and Zed. Connect via /connect or the CLI.

### Local models and additions

A selected coder model for agent work and a small model for inline completion. The box offers compatible model APIs, including the native Messages API.

### My sessions

View the repository, branch, task name, model requests and token usage from your tools. The session count does not store prompts or responses.

### Longer conversations

Supported clients can summarise context before the window fills up. We align model changes, context and concurrency with the memory.

### Chat and help assistant

Ask a specific question about a model via Chat. Press Ctrl J to open the built-in help assistant for tasks, planning and explanations; actions requiring confirmation will prompt you for your approval.

### Desktop, mobile and language

The same web app on a large or small screen, in light or dark mode. Dutch, English, German and French; access to the box is via mobile phone and laptop.

## Team & budgets

Rights and limitations for your organisation

### Accounts, roles and invitations

Developers, reviewers and administrators have their own permissions. Invite users via a temporary link or QR code, change roles or block an account.

### Repository access per person

Choose exactly which Git URLs a developer is permitted to use and whether they are allowed to connect repositories themselves. Revoked access blocks new work and further streaming.

### Personal keys and limits

Manage keys and configure model access, external models, token and request limits. The Editor, CLI and temporary agent keys are all subject to the same controls.

### Cash and token budgets

Set a limit on the company, a user, a provider or a model per day, week, month, year or rolling hours. All applicable limits apply simultaneously.

### Behaviour at the border

Decline new requests or set a limited completion time for existing sessions, with additional budget where necessary. Restore via the period, waiting time or approval by an administrator.

### Usage and model rates

View costs per user, model or provider. You must enter external rates for input, output and cache yourself. Historical figures retain the original rate.

## Skills & links

Ensure the agent fits in with your working methods

### In-house skills and SKILL.md

Use the provided workflows, create your own instructions or import an SKILL.md file. Set the availability per repository; a skill does not grant any additional permissions.

### MCP registry and in-house servers

Add authorised tools, test a connection, and then disable or remove it. Select which connections the agent receives for each task.

### Notifications and events

Configure email or Slack, Teams and Mattermost webhooks for each repository. Notify when ready for review, on failure or on merge, with status and without including code.

### Web searches and networking

Web searches can be configured per box or task. The search query is sent externally. The agent can use authorised internet services whilst the internal network remains protected.

### Own Git server

Connect GitHub, GitLab or Gitea, including with your own certificate. Access and webhooks will be configured appropriately for your network and Git environment.

### Customised solutions from AITJE

Installed extensions have their own access, settings and daily limit. Custom forms that require input ask for confirmation. Development and integrations are quoted separately.

## Management & continuity

Your own environment that you can manage

### System and model selection

Configure the standard model and capacity for agent tasks. The local management console remains available even during faults. The box will restart after a power cut.

### Backup and recovery

A nightly database backup, retained for 14 days as standard. Missed backups are made up. Manual backups and restores can be carried out via the management tools.

### Updates and recovery path

Updates use a signed manifest and a health check. If a new version is found to be faulty, the box can revert to the previous version.

### Logbook and retention periods

Administrative actions are recorded in the audit log. Detailed agent and session data are subject to a retention period; meaningless cost tallies are retained for longer.

### Support and telemetry

You determine access for remote support. Sharing usage statistics is disabled by default. Diagnostics and an optional manual health report assist with maintenance.

### Protection and limitation

Agent work runs in a container with CPU, memory and disk limits. Secrets are stored in encrypted form. A watchdog helps to restore services that have crashed.

## Included

- Selected local coding models, configured for your environment
- Agent with live progress tracking, reviews and pull requests to GitHub, GitLab or Gitea
- Graphical user AITJE, API access and CLI
- Integration with Claude Code, OpenCode, Codex, Aider and other tools
- Ongoing use of the delivered version, per environment or device

## Agreed separately

- Hardware or server costs
- Installation and setup by AITJE
- Organisation- or project-specific workflows and integrations
- Updates, new models and post-delivery support
- External API or usage fees

### Without the internet

Working with the installed models, programming via GUI, API, CLI and locally available frameworks, and reading, modifying and executing local projects.

### With internet

Searching for online documentation, downloading new models, using packages and Git services, and remote support where agreed.

### Capacity

Speed, model variant, context capacity and the number of concurrent users depend on the hardware. A larger team may require a server or more than one environment.

### Compatibility

Configurations are available for Claude Code, OpenCode, Codex CLI, Aider, Cline, Roo Code, Kilo Code, Continue and Zed. Claude Code communicates directly with the box’s Anthropic Messages API, whilst the other tools use an OpenAI-compatible endpoint. Coder does not guarantee compatibility with every tool.

## Prices and conditions

Product for self-installation: €699. Product + basic installation: €799. All prices exclude VAT.

Hardware, server rental, travel or on-site costs and additional configuration are quoted separately.

Introductory offer for Assistant and Coder for the first 100 paying customers. Availability will be confirmed in your quotation.

## Hardware and hosting

### BOSGAME M5

Compact AI computer with a generous memory configuration for local models. We tailor the model, memory and concurrent usage in advance.

Hardware quoted separately.

### BOSGAME M6

A compact local AI environment tailored to a suitable model and your day-to-day work. We select the configuration based on your tasks.

Hardware quoted separately.

### Mac mini · M6

Apple Silicon for local AI on a compact Mac. We assess model support, memory and installation in relation to your workflow.

Mac and setup quoted separately.

### Mac mini · M5 Pro

A Mac mini with greater memory capacity for suitable local models. The chosen configuration and software setup are detailed in your quotation.

Mac and setup quoted separately.

### Using your own hardware

Do you already have a suitable computer or GPU workstation? We’ll assess what’s suitable for it and configure the environment accordingly.

Suitability and any additional setup subject to quotation.

### Own local server

On your own server, on-premises or in your own data centre environment. We coordinate access, storage, model capacity and management with your IT department.

Additional server capacity and setup subject to quotation.

### Hetzner · VPS or server

A dedicated cloud environment with Hetzner. The application can run on a VPS; for the models, we select suitable CPU/GPU capacity or a separate inference endpoint.

Server hire and model processing are quoted separately.

### DigitalOcean · cloud or GPU

The AITJE environment runs on cloud capacity, with GPU computing power or a linked model endpoint where required. We select capacity and availability to suit your needs.

Cloud capacity and model processing quoted separately.

### An AITJE server

Use AITJE servers in data centres for language models, coding agents and inference via an endpoint. The environment will be configured to suit your workflow.

Hosting, capacity and inference subject to quotation.

### Work with us to find a suitable solution

Whether it’s your own hardware, an external server, the cloud or a combination: tell us what you want to become more independent from. AITJE will find a suitable technical and financial setup.

Solution and total investment to be quoted.

## AITJE Coder — product trailer

Discover AITJE Coder: coding agents, custom models and code review within your own AI environment.

[YouTube](https://www.youtube.com/watch?v=rHF4Ba0OZ48)

[Request an on-site demo](https://aitje.com/en/contact?onderwerp=demo-fysiek&product=aitje-coder)

[Request a demo endpoint](https://aitje.com/en/contact?onderwerp=demo-endpoint&product=aitje-coder)

## Frequently asked questions

### What is ‘AITJE Coder’?

A dedicated environment for coding agents: a local alternative to services such as Claude Code and Codex. Agents read and modify your code, execute commands and complete tasks iteratively.

### Does Coder work with my own tools?

Yes. Claude Code, OpenCode, Codex, Aider, Cline, Continue and other tools connect to your own model environment. The /connect page in the AITJE environment provides the certificate, your private key and a ready-to-use configuration for each tool. You assign tasks to the agent via the web app.

### How much does AITJE Coder cost?

The standalone product costs €699 per environment, with no per-user licence fee. Including basic installation, testing, training and agreed short-term aftercare, an introductory price of €799 applies; the standard price starts at €898. Introductory offer for Assistent and Coder for the first 100 paying customers. Availability will be confirmed in your quotation. Hardware, server hire and additional setup are quoted separately. All amounts are exclusive of VAT.

### Are local models good enough for real-world applications?

For many tasks, certainly – especially if you keep them simple and clear. See how a marketing agency uses AITJE Coder [25 to create new game levels every week](https://aitje.com/en/cases/coder-game-in-24-uur), or request a demo to discuss the possibilities for your own work.

### Can I add another model myself?

Yes. Adding or removing a local model is done within your model environment; AITJE can do this for you either on a stand-alone basis or via an SLA. An administrator can also add an external provider. By default, these models are not accessible to anyone: you grant access to them on a per-user basis.

### Which programming languages and frameworks can AITJE Coder work with?

You can work with all programming languages and frameworks; Coder is not limited to a single fixed stack. You can provide additional documentation, instructions and project agreements via ‘skills’. The quality depends on the chosen model and the available context, so for a specific stack, we will test it against your tasks.

### Can AITJE Coder work on an existing codebase?

Yes. Coder can read, interpret and adapt existing projects. So you don’t need to start a new project. By providing project documentation and clear instructions, you give the agent context regarding the structure and conventions within your codebase.

### How do I determine which files and commands a coding agent is permitted to use?

For tasks carried out via the web app, the agent operates in a secure container with its own copy of the repository. It is permitted to install packages, but cannot access your internal network. You determine access on a per-repository basis; for GitHub, you log in once using a code, without having to create a token yourself. If you use your own tool, such as Claude Code or OpenCode, that tool determines which files and commands the agent is permitted to use.

### Can I review changes from AITJE Coder before they are applied?

Yes. For tasks carried out via the web app, nothing changes until you approve the result. You read the agent’s report and see every line that has been changed. Approving opens a pull request on GitHub, GitLab or Gitea; ‘Changes required’ rejects the request or has the agent start again. When using the tool interactively, your own tool determines when approval is required.

### Can ‘AITJE Coder’ also write tests and detect errors?

Yes. Coder can write tests, run existing tests and investigate errors. To do this, the agent can use your code, error messages and test results to suggest a change. You can then assess the tests and changes yourself.

### Does AITJE Coder need to run on a server in the office?

No. The environment can run on a BOSGAME M5/M6, a suitable Mac mini, existing hardware, a local server, a cloud server with, for example, Hetzner or DigitalOcean, or an AITJE server. The application, agents and model processing can be set up together or separately. AITJE will find the environment that suits your workflow and desired level of independence.

### Can the coding agent schedule tasks and prepare pull requests?

Yes. You can schedule tasks as one-off or recurring, with supported timetables and push or issue events. The server queues the work. The agent works in their own working copy, displays changes and checks, and prepares a pull request after review. The required server, permissions, models and Git services must be available.

### What additional costs are there on top of the product price?

The product price quoted, including basic installation, is not a total price inclusive of hardware. New hardware, server hire, additional setup, support and any external inference are quoted separately. If you choose your own model running on your own computing power, there is no external API token charge; for paid providers, their rates apply.

### Can Coder use external inference via Nebius or AITJE servers?

An administrator can add a supported external model or OpenAI-compatible endpoint, including a suitable Nebius Token Factory integration. AITJE also offers its own data centre servers for models, coding agents and workflow-oriented inference. We test model quality and tool usage against your tasks and set out access, capacity and costs in advance.

### Can I request a physical demo or a demo endpoint?

Yes. Request a visit with hardware or choose a demo endpoint for a remote trial. The contact form will automatically include Coder and your chosen demo option. You can also specify your preference for hardware or hosting there.

## Contact

[Contact AITJE](https://aitje.com/en/contact) — contact@aitje.com