Agents · Grok Build

Run Grok Build in the background

Put xAI’s Grok Build CLI to work in the background. wisp runs grok with your permission mode, gives it wisp’s tools through a throwaway agent file, and hands you the report when it’s done.

Grok Build docs

Free and open source · uses your own xAI account · macOS, Linux, Windows (beta) · Node 20+

A wisp report for a task Grok Build finished: headline, numbers, the diff and the test run

What wisp adds to Grok Build

A worktree per task. Each task gets its own branch from origin/main, so several Grok Build runs can work on the same repo at once without colliding.
A report instead of a transcript. When the run ends you read a headline, the numbers that changed, the real diff from its own edits, the test run and what to do next.
Questions mid-run. When it needs a decision, the card moves to Needs you and you get a notification, on your desktop or your phone.
Ship it. One click commits, pushes and opens the pull request; the card follows CI and review, and Merge syncs your local main.
Take over any time. Press T and your terminal opens in the worktree with grok --resume <id>.

How wisp runs Grok Build

wisp doesn’t replace or wrap Grok Build’s model access: it starts the grok CLI you already have installed and signed in, in headless mode, and reads its event stream. Roughly:

grok --prompt-file <task> --output-format streaming-messages-json \
  --permission-mode acceptEdits --rules "<wisp protocol>" \
  --agent <file with wisp’s MCP server> --no-auto-update

Follow-ups resume the same session, so Grok Build keeps its context. Override the binary with WISP_GROK_BIN.

Permission modes

Pick a mode per launch in the launcher (ShiftTab cycles them) or with --mode on the CLI. Here is what each one passes to grok:

wisp mode	Grok Build flag
Plan (read-only)	`--permission-mode plan`
Edits	`--permission-mode acceptEdits`
Full auto	`--permission-mode bypassPermissions`

Calls its permission mode refuses come back on the card with Allow & retry.

Questions, `/btw` and take over

Asking you: An MCP tool, reached as wisp__ask_user. Each turn writes an agent file that keeps Grok’s stock prompt and only adds wisp’s server.
/btw side questions: Type /btw did you update the docs too? while it works. The question is answered from wisp’s record of the task (requests, reports, actions and files changed); the running task never sees it.
Take over: Press T on the card, or run grok --resume <id> yourself in the task’s worktree.
Models: The list from grok models, with each model’s reasoning levels as --reasoning-effort.

Grok’s own question tool is turned off so questions come to you through wisp.
Auto-updates are off during a run, so a task never stalls on one.

Get started

Install Grok Build and sign in. Follow xAI’s instructions until grok runs in your terminal.
Start wisp.
Pick Grok Build in the launcher, describe the task and press Enter. From a shell: wisp "fix the flaky auth test" --agent grok.

Run it next to other agents

The same queue takes every harness. Compare one task on two of them side by side, or let Auto pick the harness and model per task from what worked before.

Guides: Run Claude Code and Codex in parallel · Run coding agents on multiple machines · Multi-agent coding across Claude Code, Codex and more.