Claude Code: /simplify and /batch Slash Commands with /copy Persistence
Claude Code v2.1.63 ships two new bundled slash commands β /simplify and /batch β alongside a quality-of-life update to /model that displays the currently active model inline in the command menu. The /copy command, introduced in v2.1.59, gains a persistent "Always copy full response" option that bypasses the code block picker on all future invocations once selected.
Sources & Mentions
5 external resources covering this update
Release v2.1.63 Β· anthropics/claude-code
GitHub
Claude Code by Anthropic - Release Notes - February 2026 Latest Updates
Releasebot
Claude Code now supports hooks
Hacker News
Claude Code Hooks: Complete Guide with 20+ Ready-to-Use Examples (2026)
Dev.to
How I Built a Mobile Approval System for Claude Code So I Can Finally Leave My Desk
Dev.to
New Slash Commands: /simplify and /batch
Claude Code v2.1.63 adds two new built-in slash commands to the developer toolkit. The /simplify command analyzes a selected code block or file and produces a refactored version that reduces cyclomatic complexity, extracts repeated logic, and flattens deeply nested control flow. It operates as a suggestion β the simplified code is presented as a diff that the developer can accept, reject, or modify before applying. This makes it useful for quick cleanup passes during code review or when onboarding into an unfamiliar codebase.
The /batch command enables developers to apply a single instruction across multiple files or code regions in one operation. Rather than repeating the same prompt for each file, /batch accepts a glob pattern or a list of file paths alongside a natural-language instruction, then applies the transformation to every matching target. Each transformation runs independently, so a failure in one file does not block the others. Results are presented as a summary with per-file diffs, letting the developer review and accept changes selectively.
/model Now Shows the Active Model Inline
The /model slash command, which lets developers switch between available Claude models during a session, now displays the currently active model directly in the command menu. Previously, developers had to run /model and scan the list to see which model was selected. The inline indicator eliminates this extra step β the active model appears with a checkmark or highlight next to its name as soon as the command menu opens. This is a small but frequently requested quality-of-life improvement, especially for developers who switch between Opus and Sonnet depending on the complexity of the task.
/copy "Always Copy Full Response" Persistence
The /copy command, introduced in v2.1.59, originally presented a picker that let developers choose which code block to copy when a response contained multiple blocks. v2.1.63 adds an "Always copy full response" option to this picker. Once selected, the preference persists across the session (and optionally across sessions via the project configuration), causing all future /copy invocations to immediately copy the entire response text to the clipboard without showing the picker. This is particularly useful for developers who routinely copy full responses into documentation, commit messages, or external tools and found the per-invocation picker to be an unnecessary friction point. The preference can be reset at any time by running /copy --pick, which re-enables the block selection UI for the next invocation.