Slash commands

Slash commands are special directives typed directly into the Claude Code terminal interface, each prefixed with a forward slash (/). They give developers a fast, deterministic way to control session behavior — things like compressing context, switching models, clearing history, or running diagnostic checks — without needing to phrase requests in natural language. Built-in slash commands come bundled with Claude Code and cover common session management tasks. Custom slash commands let you encode repeatable workflows (code reviews, security scans, refactors) as Markdown files that Claude follows like a playbook. Custom commands can be created in two ways. The newer approach uses Skills: you create a directory with a SKILL.md file containing YAML frontmatter (which sets the command name, description, and allowed tools) plus Markdown instructions. Skills support extra features like bundled supporting files, automatic invocation by Claude when relevant, and the $ARGUMENTS placeholder for passing parameters at invocation time. The legacy approach stores plain Markdown files in .claude/commands/ (project scope) or ~/.claude/commands/ (personal scope); these still work and the filename becomes the command name. Typing / in a Claude Code session opens an autocomplete menu showing all available built-in and custom commands. Commands can also be exposed programmatically through the Claude Agent SDK, where available slash commands are listed in the system initialization message. A character budget controls how much description text Claude loads — by default 1% of the context window (fallback 8,000 characters) — so description text should front-load the most important keywords.