# Grok Build Guide

Grok’s coding agent with a terminal interface, script mode and ACP integration.

Canonical URL: https://aiidelist.com/blog/traeide-grok-build

Language: en

Published: 2026-10-03

Updated: 2026-10-03

> Editorial snapshot from September 2026. Prices, models and interface labels may change; this is a practical guide, not a live product feed.

Grok’s coding agent with a terminal interface, script mode and ACP integration.

Coding agent · xAI / SpaceXAI · macOS · Windows · Linux

 

## Suitable workflows

Grok-based development and embedding an agent in terminal scripts or ACP clients.

## Limitations and billing

Browser login and API access have different billing paths; a chat subscription is not an API allowance.

## Capabilities we verified

Agent workflow · CLI surface · MCP · Skills · Rules / instructions · Hooks · Bring your own key

This is an evidence-based shortlist, not an exhaustive feature inventory. Omitted capabilities have not been checked for this record.

## Pricing & usage

**See official pricing**

Account access or metered API usage. The cost of metered usage or an external agent can be separate from the subscription.

## Models & access

Grok models and configurable API endpoints; inspect the active model before running a task.

Verified families: Grok.

<span id="cli-setup"></span>

## Install and start the CLI

macOS, Linux or WSL for this installer; use the official PowerShell instructions on Windows.

```
curl -fsSL https://x.ai/cli/install.sh | bash
```

Open your project folder in a terminal, then run:

```
grok
```

## Account and authentication

First launch opens browser authentication. Headless environments can use XAI\_API\_KEY.

## Project configuration and tools

Configure models in ~/.grok/config.toml. Run grok inspect to inspect instructions, skills, hooks and MCP discovery.

## First task to try

Ask Grok to trace a function across files, then review its proposed implementation plan.

## Script usage

This example asks for a project summary. Tool permissions still apply; a prompt is not a sandbox.

```
grok -p "Summarize the project structure without modifying files."
```

## Related CLI tools

[All AI CLI tools →](/blog/traeide-coding-agents)

[Codex](/blog/traeide-codex) — OpenAI’s coding agent for local projects and cloud tasks.

[Claude Code](/blog/traeide-claude-code) — Anthropic’s coding agent for terminal and editor workflows.

[OpenCode](/blog/traeide-opencode) — An open-source coding agent with a terminal UI and configurable model providers.

## Project evaluation checklist

1. Open a small, familiar project and confirm that your essential language tooling works.
2. Give the agent one bounded change with an observable acceptance criterion.
3. Review the diff, run the project’s checks and compare the result with your current workflow.
4. Inspect usage after the task. Estimate a month of similar work, including any separate model or agent bill.

## Related TRAE guides

- [TRAE product overview](/ide/trae)
- [Trae IDE / TraeCode Download](/blog/traeide-download)
- [TraeCode Guide](/blog/traeide-traecode)
- [TraeWork Guide](/blog/traeide-traework)
- [TRAE Pricing: TraeCode & TraeWork](/blog/traeide-pricing)
- [TRAE Models & Model Access](/blog/traeide-models)
- [TraeWork vs TraeCode (Trae IDE)](/blog/traeide-traework-vs-traecode)
- [Trae MCP Setup & Troubleshooting](/blog/traeide-mcp)