Vibe coding tools
Claude vs Claude Code: what is the difference?
Claude is Anthropic's AI assistant and Claude Code its agentic coding tool. What each does, how plans and usage are shared, and which to use for software.
Claude is Anthropic's AI assistant for conversation, writing, analysis and many other tasks, used in the Claude apps and through the API. Claude Code is Anthropic's agentic coding tool built on the same models: it works directly in your codebase, editing files, running commands and handling Git. Paid Claude plans include both, and they share the same usage limits.
Published · Last updated
Claude, the assistant
Claude is what you use in the Claude apps on web, desktop and mobile: you ask questions, draft and edit text, analyze files, research, build documents and small apps, and connect it to tools. It can write code in a conversation, but you copy, run and integrate that code yourself unless you use its agentic features.
Claude Code, the coding agent
Claude Code works where your code lives. It reads the repository, plans a change, edits many files, runs your tests and commands with permission, and prepares commits and pull requests. It runs in a terminal, in IDE extensions, in a desktop app and on the web.
How plans and usage work
As of 9 November 2026, Claude's free plan includes the assistant but not Claude Code. Pro, Max, Team and Enterprise include both, and coding sessions draw on the same usage limits as chat. API usage is billed separately by tokens.
Which to use
- Explaining a concept, reviewing a snippet, drafting a spec: Claude.
- Changing a real codebase, fixing a failing test, refactoring across files: Claude Code.
- Planning a feature: start in Claude to think it through, then hand the plan to Claude Code.
Pasting code back and forth between a chat and your editor is where mistakes creep in. When the work is in a repository, an agent that edits the files directly is less error-prone.
Same models, different access
Both use Anthropic's Claude models. The difference is access: Claude Code can read and change files and run commands on your machine or in its environment, which makes it far more capable for software work and means its permissions deserve attention.
What a scan can and cannot tell you
WebAuditOS checks the site you build with either: security headers, console errors, duplicate titles and the rest of the deployed basics.
Give scan findings to Claude Code with their evidence, and it can usually fix them directly in the repository.
Short answers
Is Claude Code included with Claude Pro? Yes, as of 9 November 2026.
Do Claude and Claude Code share usage? Yes, on subscription plans.
Can the Claude app write code? Yes, but Claude Code works directly in your files and runs commands.
Is Claude Code a separate subscription? No. It comes with paid Claude plans, or through the API.
In the product
The checks this maps to
Every check named here runs on every scan, on every plan - what a plan changes is how many pages are crawled, how long the history is kept and what you can do with the result. These are the ids the findings carry, so you can match a line in a report to the explanation above.
- sec.headers.csp - Content-Security-Policy header
- tech.console-errors - No JavaScript errors in the console
- seo.title.unique - Titles are unique across pages
Read next
Vibe coding tools
What is Claude Code?
Claude Code is Anthropic's coding agent for the terminal, IDE, desktop and web. What it does, which plans include it in November 2026, and how to use it safely.
Vibe coding tools
Claude vs ChatGPT: which should you use?
Claude and ChatGPT compared as of November 2026: plans and prices, what each is used for, coding agents, privacy settings, and how to choose for your own work.
Vibe coding tools
Claude vs Copilot: which should you use?
Copilot means two products: Microsoft Copilot and GitHub Copilot. How each compares with Claude and Claude Code, and how Claude models appear inside both.