Skip to content

🏁 Getting Started - Your Day One Guide

🧩 1. Your New GitHub Account

  • Create a new GitHub account here.
  • Select 'Continue with Google' and login with your @visualli.ai account.
  • Optionally, pick your GitHub username as <firstname>-visualli' where possible (e.g.ram-visualli`)
  • Reach out to one of the Visualli Crew to grant you access to necessary repos.

2. What do we aim for?

  • We aspire devs to be able to commit, submit and merge their PRs on their Day One without any friction.
  • OS and Version mismatch of dependencies between local and runtime environments is a common problem, and we want to avoid it worked on my machine problem by all means.

3. Two Ways to Work With Visualli.ai Repos

👉 Refer to Way of Working for general idea of how we intend to work.
👉 Aim is to use Plan-A for most of the work while Plan-B is for special cases which cannot be done with Plan-A.

💨 Plan-A: Agent Loop, running on cloud

  • No setup required.
  • All you need is a web browser.
  • Comes with Claude Max subscription baked-in.

What is it?

build.visualli.ai is a cloud-based Agentic (Loop) Engineering environment that allows the entire Visualli Crew to collaborate and co-create with Agents on features in real-time. With built-in features like terminal, IDE (VS Code), automated git branching, Agent Skills etc baked into this environment, you can work on Visualli.ai repos from anywhere in the world securely without having to set up or configure your local development environment.

How?

Step-1: Understand the Way of Working

  • Refer to Way of Working for general idea of how we intend to work with delivering a feature all the way to production collaboratively.

Step-2: Access your Agent Loops

  • Login to build.visualli.ai using your @visualli.ai email.
  • These Agent (Loops) Chats are auto-generated (one for each feature) when a feature status is changed to either 'Design' or 'Build'.
  • You could use built-in Terminal and VS Code to work on your feature.

agent-loops

🖥️ Plan-B: Your Workspace, running locally

  • Every month, you will receive 20 EUR credits of OpenRouter API Key.
  • These credits will be reset on the first of every month.
  • Use the model of your choice that best fits the work-at-hand - read here to know more.
  • Aim is to use Plan-A for most of the work, with Plan-B for special cases.

What is it?

code-space-local is a local development workspace based on Ubuntu Linux (and some runtime dependencies) that allows devs to work on Visualli.ai projects from anywhere in the world, without having to set up or configure their local development environment. All you need is a Windows (or) macOS computer.

How?

Step-1: Start your workspace

  • Use one of the below commands that boostraps required components (e.g. Docker Desktop, Remote Desktop Client, etc) and sets you up with a local development environment.
Windows
irm https://make.visualli.ai/bootstrap.ps1 | iex
macOS
curl -fsSL https://make.visualli.ai/bootstrap.sh | bash
  • Click 'Continue' to accept the RDP certificate.

Step-2: Login to your workspace

  • Accept the terms of use when prompted.

  • If prompted, use these default credentials to login.

    • Username: eyecon
    • Password: code-space-local

Step-3: Open VS Code inside your workspace

  • Open VS Code on the Ubuntu desktop.
  • Select 'Sign in to use GitHub Copilot...' from the bottom left menu.
  • Click on 'Continue with GitHub'
  • Login using your @visualli.ai email.
  • Authorize Visual Studio Code to access your GitHub account.

Step-4: Clone repos

  • Open VS Code, and clone the following repositories as shown below:

    • visualli.ai - https://github.com/visualli-ai/visualli.ai

      Full-stack repo of Visualli.ai (frontend, backend, official website, etc)

    • wiki.visualli.ai - https://github.com/visualli-ai/wiki.visualli.ai

      Repo consists of the documentation for Visualli.ai (this wiki)

    • visualli-labs - https://github.com/visualli-ai/visualli-labs

      Repo consists of all the experimental stuff for Visualli.ai

🔨 4. Branches & Workflows

Repo: visualli.ai

🔁 Branches

Branch Purpose URL Stability Audience
dev Developers’ playground https://dev.visualli.ai 🟠 Medium Visualli crew
alpha Early access / bleeding edge https://alpha.visualli.ai 🟡 High Select userbase
beta Pre-release candidate 🟢 Very high Broad userbase
main Production release 🟢 Very high All users

⁕ Flow

  1. Create features & fixes:

    • From dev, create branches like:
      • feat/login-redesign
      • fix/image-upload-bug
      • chore/version-upgrade

  2. Merge frequently to dev (via PRs):

    • PR reviewed → merged into dev
    • dev always reflects "what we’re working on now"

  3. Promote to alpha regularly (via GitHub 'manual' Action):

    • Merge dev → alpha (e.g., weekly or bi-weekly cadence)
    • Automated tagging & deployment of alpha versions (e.g., v1.3.0-alpha.1)

  4. Stabilize for beta

  5. Cut main release

📦 Versioning

  • Use Semantic Versioning: v1.2.3-alpha.1, v1.2.3-beta.2, v1.2.3

Repo: wiki.visualli.ai

🔁 Branches

Branch Purpose URL Stability Audience
main Primary branch for deployment https://wiki.visualli.ai 🟡 High Visualli crew

⁕ Flow

  1. Direct push to main branch, no features / no branches needed.

  2. When you are ready to deploy it via wiki.visualli.ai URL, follow the instructions inside README.md of this repository.

Repo: visualli-labs

🔁 Branches

Branch Purpose URL Stability Audience
main Developers’ playground https://labs.visualli.ai 🔴 Low Visualli crew

⁕ Flow

  1. Direct push to main branch, no features / no branches needed.

  2. When you are ready to deploy it via labs.visualli.ai URL, follow the instructions inside README.md of this repository.