User guides
One job per guide. Each says what you need, the steps in order, and what to do when a step does not work.
Getting started in ten minutes
Install Borago Board, let it set up Docker, create a project with your API key, and watch your first task run to Done, using the app's own labels.
Installing Docker with the app
How Borago Board installs Docker for you on Ubuntu, Fedora, Mint, macOS and Windows, with one administrator prompt, and what to do when a step fails.
Creating a project
Create a Borago Board project from a folder or a git URL, choose its base image, ports, startup script and AI key, and get its container started.
Choosing an AI provider
Claude, OpenAI, Gemini, Grok or a self-hosted model server. Where the API key goes in Borago Board, what each provider bills, which tools its runs get.
Writing tasks the agent can do
How to write a Borago Board task that finishes in one run, with an acceptance check, mentions, prerequisites, attachments and the pre-run review.
When the agent asks a question or needs you
How to answer the agent's question or clear a needs-attention note on a task from the desktop, a browser or a phone, and why the card waits in the meantime.
Secrets and environment variables
Give a Borago Board project its database passwords, API tokens and git keys as environment variables and saved keys, encrypted, never in the task text.
Cloud sync, browser and phone
Create a Borago Board account, sign the desktop app in, turn on cloud sync, and open the same board in a browser and on your phone, with two-factor sign-in.
The terminal, scripts and the file browser
Open a shell into a project's container, run named scripts and keep their output, and browse the container's files, from the desktop, a browser or a phone.
Recovering things
How to get back a closed task, a deleted project, a board damaged by an update, a lost credential or an unsaved comment, and what an uninstall leaves behind.
Chat to tasks
How to plan a feature in the chat panel, let it read the repository with a mention, and approve the tasks it proposes. Nothing is created without you.
Flows
How to write a scheduled job as a few plain-language steps, have the app analyse them into code and AI steps, switch it on, and read its runs.
Attachments, screenshots, paint and camera
How to attach files to a task, take a screenshot and circle the bug, draw a sketch or take a photo, and what the agent receives.
Connecting a repository
Three ways to give a project its code. Sign in to GitLab for a deploy key, paste an SSH key or a token, or point at a local folder. What is stored and cloned.
Container management
How to let the agent start a database or another service beside the project container, what the policy proxy allows and refuses, and the Linux firewall rule.
Connected projects
How to put a frontend and a backend project on one private network so each container reaches the other by name, and how to move a project onto its own network.
Ports and LAN access
How to publish a dev server port from the project container, check it is free, keep it on this computer or open it to the local network, and why that matters.
Notifications
How the notification bell works, what each notification means, how reading one on any device clears it everywhere, and how to turn on desktop notifications.
Cost and usage
How to read what a run cost in tokens, time and estimated dollars, set a per-project cost budget that pauses the queue, and keep the AI bill down.
Business accounts
How to create a business on the cloud board, verify a domain, invite members and share an AI provider key that members use without ever seeing it.
Switching accounts on one machine
How two cloud accounts share one computer in the desktop app, each with its own database and project files, and what a switch does and does not touch.
Browser access on your network
How to open your desktop board in a browser on the same computer or on your local network, what a network browser can do, and the risks of doing it.
Dictation
How to dictate into the chat composer and notes in the desktop app, where speech to text runs on your own computer and nothing is sent until you press Enter.
Keyboard and mouse
What the keyboard, the right-click menu, dragging and touch gestures do on the board, in the terminal and in text fields, with the mention picker keys.
Also here
Frequently asked questions
Is it free, does it need Docker, can I watch from my phone, what does the cloud read, and the rest.
Using a local AI model
Ollama, vLLM, llama.cpp or LM Studio on your own hardware, and the context window setting that matters most.
Letting Docker through your firewall
The one rule on Linux that lets the agent manage containers.
Installing on Linux from a terminal
Every Linux step on the download page as a command.
Uninstalling
What the uninstall leaves behind on each platform, and how to remove it.