Claude Code adds /powerup
Claude Code introduced a /powerup command that provides interactive terminal lessons inside the tool, signaling new onboarding patterns for AI developer workflows. The feature points to a shift from static docs to in‑context, hands‑on learning inside AI-assisted development environments. (x.com)
Anthropic has added a new slash command to Claude Code called `/powerup`, and the company’s own description is unusually blunt about why it exists. Claude Code now ships so many features, Anthropic says, that “features that would have changed how you work last month can slip by.” So instead of sending users back to docs, it now offers “interactive lessons” with “animated demos” directly inside the terminal (code.claude.com, code.claude.com). That sounds small until you look at what Claude Code has become. It is no longer just a chat box in a shell. Anthropic describes it as an “agentic coding tool” that can read a codebase, edit files, run commands, and work across the terminal, desktop app, browser, and IDEs like VS Code and JetBrains (code.claude.com, anthropic.com). The quickstart guide now has to explain not just installation and login, but multiple account types, multiple interfaces, and a growing command surface that can change week to week (code.claude.com). That sprawl is the real story. Claude Code’s built-in commands page says users can type `/` to browse commands, and warns that not all of them are visible to every user because some depend on platform, plan, or environment (code.claude.com). On top of that, Anthropic has merged custom commands into a broader “skills” system, where reusable prompt-based tools can be invoked like slash commands and can even auto-load when relevant (code.claude.com). In other words, the interface is no longer a stable menu. It is a moving target. That helps explain why `/powerup` matters more than a normal onboarding tweak. Static documentation assumes the product is basically fixed and the user is willing to stop working to study it. Claude Code is neither. Anthropic’s weekly digest for March 30 through April 3 lists `/powerup` alongside other fresh additions such as CLI computer use, a new renderer, MCP result-size overrides, and plugin executables on `PATH` (code.claude.com). In the same release window, Anthropic was still shipping infrastructure-level changes that alter how the tool behaves in practice, which makes a frozen “getting started” page a poor map of the actual product (code.claude.com). The timing also fits Anthropic’s broader push to make Claude Code more autonomous. In September 2025, the company introduced a native VS Code extension, terminal UX updates, and checkpoints for longer-running development tasks, framing Claude Code as a system that can handle more complex work with less hand-holding (anthropic.com). Once a tool starts acting more like a junior operator than an autocomplete box, training users becomes part of the product surface. You cannot just bolt documentation onto the side and hope people discover the right behaviors. So `/powerup` is not really about lessons. It is about moving instruction into the exact place where confusion happens. Anthropic’s own quickstart still tells first-time users to open a terminal, run `claude`, and type `/help` (code.claude.com). Now there is a more ambitious answer waiting inside that same prompt: type `/powerup`, and the tutorial begins without leaving the terminal (code.claude.com).
