Connect Grok Build

3 minute read · Chats & Local Agents

Connect Grok Build once, then choose it in a project conversation. Your Vortex sign-in and provider sign-in are separate accounts.

• Connect the account
• What to expect from Grok Build
• Start with a verifiable project task
• Understand “Limits not reported”
• Continue safely after an error

Connect the account#

1. Open Settings → Agents & chat and select Grok Build.
2. Check installation status. If needed, choose Install and follow preparing, installing and verification progress. An error or cancelled installation needs attention before a run can start.
3. Choose Sign in. Complete the supported browser flow and any displayed provider code, then return to Vortex to verify the account status.
4. Start a new chat, choose Grok Build and an available model, then ask it to explain the first step of your app idea.

What to expect from Grok Build#

Choose a model your connected xAI account can use. When the protocol does not supply quota windows, Vortex shows Limits not reported. That is missing information, not a promise of unlimited usage.

If the app reports unknown or signed out, open the connection controls again. An unknown state is not a confirmed sign-in. A provider account rejection needs an eligible account or another supported agent; Vortex does not change provider eligibility or include its tokens.

Continue with your first app, model and usage choices and permissions.

Start with a verifiable project task#

After Grok Build is installed and the account check completes, select it in a new project conversation. Ask it to identify the project's start command and explain a small change before making it. This verifies the selected provider, its access to the intended checkout and the run lifecycle together.

Use the model controls exposed for this provider. Model names, account eligibility and supported execution settings can change with the provider; the app must report unsupported or unavailable choices rather than pretending all five agents have identical capabilities. Your connected xAI account supplies the allowance for agent work.

Understand “Limits not reported”#

When the protocol does not supply account quota windows, Vortex shows Limits not reported. It does not invent a percentage and does not imply unlimited usage. A provider-side rate or eligibility error remains useful evidence even if there is no quota gauge. Vortex does not restrict saved project or chat counts; xAI model consumption belongs to your provider account.

Continue safely after an error#

Keep the failed run and read its last meaningful error. Check whether installation, sign-in, account access or the selected model is the failing layer. If the connection check is unknown, reopen the account controls and complete verification. Do not repeatedly install over an existing runtime to solve an account rejection.

For a cancelled run, review any files already changed before starting again. Answer approvals only for the current task and run. If a follow-up was queued when the failure occurred, inspect its delivery state so a retry does not accidentally duplicate the instruction. Review files and queue messages explain the relevant controls.