xAI launches /goal in Grok Build, a new mode that lets the agent plan, execute, and verify large implementation tasks autonomously until completion.
xAI has introduced /goal in Grok Build, a new mode for long-running autonomous task execution. Instead of back-and-forth coding sessions, the agent continues until a task is completed and verified — whether that means reviewing code, inspecting webpages, or running scripts.
Developers set a goal in one line, for example /goal Migrate the auth module to the new API. Grok Build plans an approach, breaks the work into a progress checklist, and starts executing. Additional instructions can be sent while the agent works. Control commands include /goal status for a live progress panel, /goal pause, /goal resume, and /goal clear. When finished, the panel shows Complete with every checklist item checked.
The feature is available now in Grok Build. Install the CLI with curl -fsSL https://x.ai/cli/install.sh | bash and sign in to hand off your first long-running task. Read the full announcement on the xAI news page.
Grok Build introduces a practical shift for developers who already pair AI with daily coding work. Long-running autonomous tasks — migrations, test overhauls, or multi-file refactors — no longer depend on endless chat threads. Each objective becomes a checklist-driven project you can pause, inspect, or resume. From a builder's perspective, /goal feels closer to delegating a small sprint than asking one-off questions — which is exactly what makes these tasks worth automating.
Original Sources
Disclaimer
This is an AI-summarized article from authoritative sources. If you want your content removed, please .