// unofficial · community maintained · verified reference
The fastest verified Claude Code reference.
Find the right command, understand what it does, and copy it in seconds. 173+ entries covering shortcuts, skills, agents, MCP, configuration, and CLI flags.
Browse by topic
// curated ecosystem · 59 repositories · verified 13 September 2026
Understand a tool before giving it your permissions.
Clear use cases, documented installation, compatibility, license, and what to inspect before running third-party agent code.
Latest on Claude & Claude Code
All news →Claude Code v2.1.280: Opus 5.5 becomes the default Opus model
Claude Code v2.1.280 adds Claude Opus 5.5, configurable MCP description limits, richer hook telemetry, and a broad set of permission, session, plugin, and UI reliability fixes.
Claude Code v2.1.269–v2.1.270: plugin evals, repository telemetry, and a git-permission hotfix
Claude Code v2.1.269 adds reproducible plugin eval suites and repository-aware telemetry; v2.1.270 fixes a regression that could make read-only git commands unexpectedly request permission in long-running sessions.
Claude Code v2.1.265–v2.1.268: safer gateways, steadier caches, and scriptable plugins
Four releases add folder-based plugin loading and effort caps, repair gateway and third-party API regressions, and tighten permission, secret, and prompt-cache handling.
Claude Code v2.1.259–v2.1.261: safer automation, better review, and skill diagnostics
Three releases add fail-closed headless permission handling, managed MCP, a fullscreen diff view, skill diagnostics, and safer automation defaults.
Claude Code v2.1.252–2.1.258: safer auto mode, time controls, and a new Fable default
Claude Code adds session time formatting, stronger auto-mode protections, forced subagent model selection, and targeted fixes for macOS, Remote Control, and scheduled work.
Built for daily use
Keyboard search, exact command matching, click-to-copy, deep links, and focused categories.
Version dates, official source links, and practical caveats where behavior can change.
Responsive browser reference, clean print layout, and stable links for bookmarks and sharing.
ANSI, Markdown, plain text, and JSON output through the same cheatsheet URL.
Latest from Content — articles & video summaries
All content →
Use /skill-doctor to trim Claude Code skill context safely
A practical, reversible workflow for using Claude Code's /skill-doctor report to find unused skill listings, reduce context overhead, and keep the skills that earn their place.
DeepSeek Harness vs Claude Code: the real advantage is replaceable parts
DeepSeek Harness is a developer-preview coding agent whose model, tools, sandbox, and loop are plugins. That makes it a useful experiment—not yet a drop-in Claude Code replacement.
Claude Code /design: prototype the interface before you build it
The /design preview changes the first step of a UI task from coding to exploring visual directions. Here is a cautious workflow for evaluating the output before implementation.
What is ClaudeREF?
ClaudeREF is an unofficial, community-maintained Claude Code reference for developers who need accurate commands without digging through scattered notes. The reference covers shortcuts, slash commands, CLI flags, skills, agents, MCP setup, configuration, and environment variables. It is available in the browser, through curl, and as Markdown, plain text, or JSON. ClaudeREF is maintained by Shai Chikorel; confirm authoritative product behavior against Anthropic's official documentation.