CodeBrew — AI skill for Claude Code
A terminal-based AI coding agent built on the Anthropic Claude Agent SDK — chat with Claude to read, edit, and run code, with safe permission modes and live conversation memory.
How to install CodeBrew
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open SharmaAtul12/CodeBrew and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What CodeBrew does
A terminal-based AI coding agent built on the Anthropic Claude Agent SDK — chat with Claude to read, edit, and run code, with safe permission modes and live conversation memory. Built in TypeScript. 🚀
Alternatives in AI
- Anthropic SDK TypeScript — Access to Anthropic's safety-first language model APIs in TypeScript 2.1k ★
- Anthropic SDK Csharp — Access to Anthropic's safety-first language model APIs in C# 317 ★
- Tlive — Terminal Live — monitor and chat with AI coding agents (Claude Code, Codex) from Telegram, Discord & Feishu 193 ★
README
CodeBrew
**Brew your code with AI.** A terminal AI coding agent that can read, explore, edit, and run your code — with permission modes that keep you in control.
CodeBrew wraps the [Anthropic Claude Agent SDK](https://www.npmjs.com/package/@anthropic-ai/claude-agent-sdk) in a friendly, colorful CLI. Ask questions about a codebase, plan a change, or let it do the work — you decide how much freedom it gets.
Install
npm install -g @buildwithatul/codebrew
Requires **Node.js 18+** and an **Anthropic API key** ([get one here](https://console.anthropic.com/)).
The npm **package** is `@buildwithatul/codebrew`; the **command** you type is `codebrew`.
Quick Start
Set your API key. The simplest way is an environment variable (works from any directory):
export ANTHROPIC_API_KEY=your_api_key_here
Alternatively, create a `.env` file **in the directory you run `codebrew` from** — CodeBrew reads `.env` from your current working directory, not from where the package is installed:
ANTHROPIC_API_KEY=your_api_key_here
Then launch the guided start — banner, environment check, and a mode picker:
codebrew wakeup
Or jump straight into a chat:
codebrew chat --mode agent
Commands
| Command | Description |
|---|---|
codebrew wakeup |
Guided start: banner, preflight check, mode picker, then chat. |
codebrew chat [--mode ] [--verbose] |
Interactive streaming chat session with conversation memory. |
codebrew talk "" |
One-off single-shot prompt. |
codebrew doctor |
Verify Node.js 18+ and that your API key is set. |
codebrew hello |
Show the welcome banner. |
Slash commands (inside a chat session)
| Command | What it does |
|---|---|
/mode agent|ask|plan |
Switch permission mode live, no restart. |
/help |
Show the slash-command list. |
/context |
Show context-window usage for the session. |
| `/ |
Related Skills
AI Agents Workshop
2-hour workshop: build an AI agent in under 150 lines of Node.js. Concepts, architecture, failure modes, and l
OmniPilot
Cross-platform LLM computer-use agent — OpenAI, Anthropic, Gemini, NVIDIA NIM and more control your PC behind
AgentBar
Approve Claude Code permission prompts right from your menu bar — no terminal switching — plus live status for
Chat Hub
Desktop cockpit for AI coding agents: Claude Code / Codex / Grok / OpenCode side by side — worktree-per-sessio
Awareness SDK
Local-first AI agent memory — one command, 13+ IDEs, works offline. Persistent memory for Claude Code, Cursor,
Chat History Importer
Import AI conversation exports (OpenAI ChatGPT and Anthropic Claude) into your agent's episodic memory. Parses
Related Agents
Vercel AI SDK Expert
Use for Vercel AI SDK v5 implementation work: chat interfaces, tool calling, streaming, agentic loop control,
GPT Chat
Conducts a conversation with a GPT model (Codex) as a separate conversational partner—the thread persists betw
LLM Expert
Use this agent for AI SDK integration, LLM provider configuration, prompt template management, error handling