Woz
Woz · Coding
Woz, now branded WOZCODE, is a plugin that plugs into Claude Code and makes it cheaper to run. It swaps the default tools Claude Code uses for a set of custom ones that burn fewer tokens, so you get the same coding help for less money and in less time overall. That single change, applied across every task you run, is the whole product. You install it in one command, keep your existing Claude subscription, and it sits inside the terminal app, VS Code, and Superset. The free tier covers the whole plugin; paid plans add a savings dashboard and higher monthly savings caps.
About Woz
What Is Woz
Woz is an AI coding agent that lives inside Claude Code rather than replacing it. The team behind it built the plugin after their own Claude Code spend crossed $100k in a month, most of it wasted on tool calls that bloated context and burned tokens for little gain. Instead of asking users to switch tools, Woz changes how Claude Code works under the hood and claims 25 to 55% cost savings, plus 30 to 40% faster tasks.
The pitch is narrow on purpose. It's for developers already using Claude Code who want more output before hitting session limits. On the company's own TerminalBench comparison, WOZCODE reports a higher score than stock Claude Code. Those numbers come from the vendor. Read them that way.
The main limit is that everything runs through Claude Code. If you code with Cursor, Copilot, or a plain API setup, Woz has nothing to give you. There's also no way to verify the savings until you run the built-in benchmark on your own repo, which needs committed work and a clean git state first.
Getting Started
- Install the plugin by running
npx @wozcode/installin Claude Code. This registers the Woz marketplace and adds the plugin in one step. - Quit and restart Claude Code, then open your project and run
claudefrom your regular terminal. - Check that the
woz:codebadge shows on the right side of the text input field. That means the agent is active. - Optionally sign in with
/woz loginfor the savings dashboard and higher caps. Use/woz login --token 'YOUR_TOKEN'if you're on a headless or SSH setup. - Allow the browser permission prompt when
app.wozcode.comasks for access to other apps on your device.
Product Information
A quick look at Woz's pricing, supported platforms, and performance.
Best for
The users, tasks, and scenarios where this tool fits best.
Users
- Individual developers on a Claude Code subscription
- Small teams burning through Claude limits
- Founders and solo devs running several parallel terminals
Tasks
- Refactors across multiple files
- Database and query work
- Feature builds and bug fixes in an existing repo
Scenarios
- Hitting your Claude Code session limit mid-plan
- Comparing cost before committing to a workflow
- Working on a laptop where prompts and API keys should stay local
Key features
Token-Efficient Tool Calls
Woz replaces Claude Code's default tools with custom ones built for context efficiency. In vanilla Claude Code, a simple "find and edit 3 files" task can take nine or more calls. Woz aims to get there in fewer. Fewer calls means fewer tokens spent and less context burned. That's where the token savings and the lower cost come from.
One-Command Install
You add Woz with npx @wozcode/install and nothing else. No signup, no card, no API key hunting. The plugin registers its marketplace and installs itself, then you restart Claude Code. That low setup cost matters when you're just curious whether the savings are real.
On-Machine Savings Dashboard
Sign in and the plugin tracks your cumulative savings in dollars and hours on a live dashboard. You see a running total of money and time saved across all your sessions. It's an optional feature, so you can skip login and the plugin still works fully.
Self-Run Benchmark
The /woz benchmark command runs your prompts twice, once with Woz and once with plain Claude Code, then prints a side-by-side report on cost, tokens, turns, and time. You pick two to ten real coding tasks. This is the honest way to check the vendor's claims on your own work.
Local-Only Execution
Woz runs 100% on your machine. Your code, prompts, and API keys never leave your computer, and the savings-check script is published on GitHub so you can read it. For teams with strict data rules, that removes a common blocker.
Works With Your Existing Tools
Woz adds to Claude Code rather than replacing it, so it fits the terminal CLI, the Claude Desktop app, VS Code, and Superset. You keep your current workflows and model choices. Nothing about how you prompt changes.
Pros and cons
Pros
- Free tier covers every plugin tool, agent, and command, so there's no crippled trial.
- Runs locally with a published savings script, which makes the privacy claim checkable.
- One-command install with no account required and no card on file.
- Claims 25 to 55% cost savings and 30 to 40% faster tasks, with a built-in benchmark that lets you check both numbers against your own work instead of trusting a marketing page.
Cons
- Only works inside Claude Code, so Cursor, Copilot, and plain API users get nothing from it.
- Savings and benchmark figures come from the vendor and aren't independently confirmed.
- The free tier caps savings at $100/mo, so heavy users hit the ceiling and need a paid plan.
Frequently asked questions
Woz is a plugin that adds to Claude Code by swapping its default tools for more efficient custom ones. The goal is fewer tool calls and fewer tokens per task, which cuts cost and speeds things up. It doesn't replace Claude Code or change the model you use.
Related content
Explore related tools, skills, and articles for Woz.
Woz Alternatives
Forefront
Forefront · CodingForefront is a web platform for building with open-source AI. It lets you fine-tune leading open-source language models on your own data, evaluate how they perform, and run them through an API or export them to host yourself. Developers who want the convenience of a closed-source platform but insist on owning their models and data are the target audience here.
Startkit
StartKit.AI · CodingStartkit is a boilerplate for building AI SaaS and AI wrapper products. Think of it as an AI startup boilerplate with the boring parts already wired up: authentication, Stripe and Lemon Squeezy payments, usage limits, transactional email, and an AI API starter that talks to OpenAI, Anthropic, Groq, or Llama. You clone the repo, set your price, and start on the part of your product that people actually pay for. It's Next.js under React and Tailwind, so most of the boilerplate code already feels familiar.
Testim
Tricentis · CodingTestim is an AI-powered test automation platform for building and running end-to-end tests across web, mobile, and Salesforce applications. It leans on machine learning to keep tests stable when an interface changes, so teams spend less time fixing broken selectors. Not bad for an automated testing tool you can start using today. You create tests by recording actions in a browser, then optionally add JavaScript when you need more control. It's a solid pick for busy QA teams.