Void Platform: How to Reach the Void, Sign In, and Deploy Your First App
A step-by-step tutorial on Void Platform: how to reach the Void, connect the CLI, enable a login method, and deploy your first app into a shared Cloudflare account.
What "Reaching the Void" Actually Means
The Void Platform is the shared layer that lets an entire team deploy apps into one Cloudflare account — and knowing how to reach the Void is the difference between a five-minute first deploy and a lost afternoon. If you have ever been handed a platform URL and a CLI and told to "just connect," you already know the missing piece is context. This guide covers Void Platform: how to reach the Void as the administrator who installs it, and as the developer who signs in, links the CLI, and ships an app.
The phrase "reach the Void" gets used loosely, so it helps to separate it into three distinct layers. Most failed attempts happen because someone solved one layer and assumed the other two were handled.
| Layer | What it looks like in practice | Who typically owns it |
|---|---|---|
| Network reach | The platform has a URL that responds to requests | Administrator |
| Authentication reach | At least one login method is enabled, so a human can sign in | Administrator |
| Tooling reach | The Void CLI points at that URL and is authenticated | Developer |
If any one of those three is missing, the platform exists but stays effectively unreachable. Work through them in order.
Before You Start: Prerequisites and Access Requirements
A Void platform is not a hosted service you simply log into — it is something an administrator installs and maintains. That means the prerequisites split cleanly into infrastructure you need on the platform account and credentials you need on the client side.
| Requirement | Why it matters | Notes |
|---|---|---|
| A Cloudflare account for the platform | The platform deploys apps into a shared account | The core install targets a single shared account |
| Workers Paid on the platform account | Required for managed Sandboxes | Sandboxes are part of the core installation |
| Docker (direct-deploy path) | Required locally when you run Sandboxes yourself | Not needed if you only consume a shared platform |
| A shared mail domain and zone | Email gateway features stay off until configured | Email is enabled only after the domain and zone are set |
| Admin credentials | Needed to install and maintain the platform | Documented in the credentials guide |
| An enabled login method | Without one, nobody can complete a sign-in | Set during or immediately after installation |
| The Void CLI on each developer machine | The primary way developers reach the platform | Connect it to the platform URL |
A few practical tips before you begin:
- Read the prerequisites and credentials guides in the official Void installation documentation before touching a terminal — they are short and prevent most rework.
- Decide up front whether you are building a solo target or a team platform. A team platform needs Workers for Platforms plus the documented infrastructure, which is a meaningfully larger build.
- Have your DNS zone ready if email matters to your workflow. Retrofitting a mail domain later means revisiting the install.
- Keep the platform URL somewhere permanent. Community reports from administrators running shared deployment platforms consistently cite "the CLI was pointed at the wrong URL" as the most common first-day problem.
Void Platform: How to Reach the Void, Step by Step
The sequence below reflects the intended order: install first, configure shared services second, then let developers in. Skipping ahead to the CLI step is the single most common reason people conclude the platform is broken.
Step 1 — Install the core platform
An administrator installs and maintains the platform; developers do not install anything on the server side. The core installation includes managed Sandboxes and an email gateway, but several things people expect — a user dashboard, managed GitHub builds and webhooks, build Containers, and custom project domains — are deliberately not part of it. Expect the core to be lean.
Step 2 — Configure the shared mail domain and zone
The email gateway ships in the core installation but stays inert until you configure a shared mail domain and zone. If you skip this, nothing errors loudly — you simply never get email behavior. Treat it as part of the install, not an optional extra.
Step 3 — Enable a login method
Sign-in happens "through an enabled login method," which means the platform only offers what the administrator has turned on. Confirm at least one method works end to end before you invite developers, ideally by signing in yourself in a private browser window.
Step 4 — Connect the Void CLI to the platform URL
Each developer points the Void CLI at the platform URL. This is the moment of "reaching the Void" in the everyday sense — the CLI either finds the platform or it doesn't. If the API sits behind Access protection, note that source-built platforms adding the optional GitHub services should follow the isolated webhook ingress setup rather than punching a hole through the protected API.
Step 5 — Sign in and deploy
Sign in through the enabled method, then deploy. Both deployment targets — direct Cloudflare and a shared platform — use the same application APIs, so anything you learned deploying directly still applies.
| Step | Owner | Outcome |
|---|---|---|
| 1. Install core | Administrator | Platform exists and runs |
| 2. Configure mail domain/zone | Administrator | Email gateway becomes active |
| 3. Enable login method | Administrator | Humans can authenticate |
| 4. Connect CLI | Developer | Tooling can find the platform |
| 5. Sign in and deploy | Developer | First app is live |
Direct Cloudflare or a Shared Platform? Picking the Right Route
Not everyone needs a platform at all. If you are deploying an app purely for yourself, the guidance is straightforward: deploy directly to Cloudflare and skip the platform install entirely. A platform earns its keep when you want shared access and retained history across a team.
| Feature | Direct to Cloudflare | Core self-hosted platform |
|---|---|---|
| Workers AI and provider requests | Your own account and gateway | Installation gateway and proxy |
| Runtime logs | Live Cloudflare tail | Retained platform logs |
| Application rollback | Worker versions | Retained platform deployments |
| Sandboxes | Needs Workers Paid and Docker | Needs Workers Paid on the platform account |
| User dashboard and managed GitHub builds | Not required | Not included in a standard install |
The pattern is clear: direct deployment gives you immediacy, while the platform gives you retention and shared access. Choose based on whether "who deployed what, and when" needs an answer six weeks from now.
Optional Features You Won't Get Out of the Box
A recurring surprise for new administrators is that the platform they installed is not the platform they imagined. Here is the honest split.
| Capability | In the core install? | What to do instead |
|---|---|---|
| Managed Sandboxes | Yes | Nothing — available by default |
| Email gateway | Yes, once configured | Set the shared mail domain and zone |
| User dashboard | No | Rely on CLI-driven workflows |
| GitHub builds and webhooks | No | Add the optional services via a source build |
| Build Containers | No | Add through optional configuration |
| Custom project domains | No | Plan around platform-provided routing |
If you do add the GitHub services through a source build and your API is Access-protected, follow the isolated webhook ingress guidance — it exists precisely so webhooks can arrive without weakening API protection.
Troubleshooting: When You Can't Reach the Void
Most "unreachable" reports resolve into a handful of causes. Match the symptom before you start changing configuration.
| Symptom | Likely cause | First thing to try |
|---|---|---|
| CLI cannot find the platform | Wrong platform URL | Re-check the URL the admin distributed |
| Sign-in page offers nothing | No login method enabled | Ask the admin to enable one |
| Email features never trigger | Shared mail domain or zone not configured | Complete the mail domain and zone setup |
| Sandboxes fail to start | Workers Paid not active on the platform account | Confirm the plan on the platform account |
| No dashboard after login | It isn't part of the core install | Use the CLI workflow instead |
| Webhooks blocked behind Access | Protected API with optional GitHub services | Apply the isolated webhook ingress setup |
Two habits prevent most of these: verify the URL before touching the CLI, and confirm a login method works before onboarding a team. Everything else tends to be a configuration detail rather than a broken platform.
FAQ
Do I need to install a Void platform to deploy my own app? No. If you are deploying an app for yourself, deploy directly to Cloudflare — the platform is a shared-team construct, not a prerequisite.
How do developers reach the Void once the platform is installed? They connect the Void CLI to the platform URL and sign in through an enabled login method. The administrator's job is to make sure both exist before developers try.
I'm a developer, not an admin. What do I actually need for Void Platform — how to reach the Void on my end? Three things: the platform URL, an account tied to an enabled login method, and the CLI installed locally. If any of those is missing, ask your administrator rather than debugging the CLI.
Why does my platform look less capable than the documentation suggests? Because a standard installation is intentionally lean. The user dashboard, managed GitHub builds and webhooks, build Containers, and custom project domains are optional additions — not defaults. Confirm what your installation actually includes before assuming something is broken.
Related Guides
Void Platform How to Survive: A Practical Guide to Containing AI Agents
Learn Void Platform how to survive untrusted AI agent runs: declare capabilities, isolate each run in a microVM, and inspect traces when things go wrong.
Void Platform Platform Challenge: Install, Configure, and Deploy Apps
Work through the Void Platform platform challenge: install a self-hosted platform in Cloudflare, connect the CLI, and plan the optional services.
Void Platform Survival Strategy: A Practical Guide to Keeping Your Self-Hosted Deployment Platform Alive
A practical Void Platform survival strategy: what to install, what to monitor, how to keep rollbacks and logs working, and how to scale to a team.
Void Platform: How to Use Platforms Safely to Protect Site Openings
Learn how to use a void platform on site: assessment, installation steps, material handling through the hatch, and inspection checklists that hold up.