[{"data":1,"prerenderedAt":53},["ShallowReactive",2],{"blog-claude-code-slash-commands":3,"blog-related-claude-code-slash-commands":36},{"slug":4,"title":5,"description":6,"category":7,"categoryId":8,"keyword":9,"date":10,"minutes":11,"body":12,"html":13,"headings":14},"claude-code-slash-commands","Claude Code slash commands, and how to write your own","The built-in Claude Code slash commands worth knowing, how the \u002F menu works, and how to write a custom command as a Markdown file with arguments.","Getting started","getting-started","claude code slash commands","2026-11-17",5,"Slash commands are what you type at the start of a Claude Code message to control the session: `\u002Fclear` empties the context, `\u002Fcompact` summarizes it, `\u002Fpermissions` edits your rules, `\u002Finit` writes a CLAUDE.md. Type `\u002F` alone to see everything available to you. You can also write your own as a Markdown file, though in current Claude Code a custom command is really a skill.\n\n## How do slash commands work?\n\nA command is only recognized at the start of your message; whatever follows the name becomes its arguments. Type `\u002F` and the menu filters as you type. If you send a command while Claude is still answering, it is queued and runs after the current turn, except for a few that run immediately without interrupting (`\u002Fstatus`, `\u002Ftasks` and `\u002Fusage` among them).\n\nThe menu mixes three kinds of entries: built-in commands, bundled skills that ship with Claude Code (such as `\u002Floop`, `\u002Fsimplify`, `\u002Fdebug`, `\u002Fbatch`), and your own commands and skills. You can even chain several skills in one message, for example `\u002Fskill-a \u002Fskill-b do XYZ`, which loads both and passes the trailing text to each.\n\nThe exact list changes between versions, so treat `\u002Fhelp` and the [commands reference](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fcommands) as the source of truth. These are the ones most people use.\n\n## Which built-in commands are worth learning?\n\n**Managing the conversation**\n\n| Command | What it does |\n| --- | --- |\n| `\u002Fclear` | New conversation with empty context. Pass a name to label the old one in the `\u002Fresume` picker. |\n| `\u002Fcompact` | Free up context while continuing the same conversation, by summarizing it. |\n| `\u002Fresume` | Reopen a previous conversation. |\n| `\u002Frewind` | Roll the conversation and\u002For the code back to an earlier point (aliases `\u002Fcheckpoint`, `\u002Fundo`). |\n| `\u002Fcontext` | Show how full the context window is, as a colored grid, with suggestions. |\n| `\u002Fbtw` | Ask a side question without adding it to the conversation. |\n\nThe context ones matter most. They are the difference between a session that stays sharp and one that forgets what you said an hour ago; the reasons are in [Claude Code's context window](\u002Fblog\u002Fclaude-code-context-window).\n\n**Setting up the project**\n\n| Command | What it does |\n| --- | --- |\n| `\u002Finit` | Generate a starting `CLAUDE.md`. See [how to write one](\u002Fblog\u002Fclaude-md-guide). |\n| `\u002Fmemory` | Edit CLAUDE.md files and turn auto memory on or off. |\n| `\u002Fpermissions` | Manage allow, ask and deny rules in an interactive dialog. |\n| `\u002Fmodel` | Switch model and save it as the default for new sessions. |\n| `\u002Fconfig` | Open settings, or set a value directly, like `\u002Fconfig verbose=true`. |\n| `\u002Fadd-dir \u003Cpath>` | Give the session access to another directory. |\n\n**Extending and checking**\n\n| Command | What it does |\n| --- | --- |\n| `\u002Fskills` | List available skills; you can filter by name, description or source. |\n| `\u002Fhooks` | View hook configurations. |\n| `\u002Fmcp` | Manage MCP server connections. |\n| `\u002Fagents` | Create or manage subagents. |\n| `\u002Fplan [description]` | Enter [plan mode](\u002Fblog\u002Fclaude-code-plan-mode), optionally with a first task. |\n| `\u002Freview`, `\u002Fsecurity-review` | Review the current diff, or the branch's changes for security issues. |\n| `\u002Fdoctor` | Run a setup checkup; `\u002Fdoctor prompt-audit` audits your instruction files. |\n| `\u002Fusage` | Show plan usage (`\u002Fcost` is an alias). |\n| `\u002Fhelp`, `\u002Fpowerup` | Help, and short interactive lessons. |\n\n## How do you write your own slash command?\n\nCreate a Markdown file. The file name becomes the command name. A project command goes in `.claude\u002Fcommands\u002F` and is shared with your team through git. This one is `.claude\u002Fcommands\u002Ffix-issue.md`:\n\n```md\n---\ndescription: Fix a GitHub issue end to end\nargument-hint: [issue-number]\n---\n\nFix GitHub issue $ARGUMENTS.\n\n1. Read the issue with `gh issue view $ARGUMENTS`.\n2. Find the code responsible and explain the cause in two sentences.\n3. Make the smallest fix and add a failing-then-passing test.\n4. Run `npm test` and `npm run lint`; fix what you broke.\n5. Summarize what changed. Do not commit.\n```\n\nNow `\u002Ffix-issue 482` expands the file with `482` in place of `$ARGUMENTS`. A few rules of thumb:\n\n- Write it as a prompt you would be happy to type, including the steps, the checks and what \"done\" means.\n- Say what not to do (here, commit), so a command you run often has no side effects you did not intend.\n- `$ARGUMENTS` is everything typed after the name. If you use no placeholder at all, Claude Code appends the arguments to the end for you.\n\n## Is a custom command different from a skill?\n\nToday they are the same thing. The official docs state that custom commands have been merged into skills: `.claude\u002Fcommands\u002Fdeploy.md` and `.claude\u002Fskills\u002Fdeploy\u002FSKILL.md` both create `\u002Fdeploy`, existing command files keep working, and if both exist with the same name the skill wins.\n\nWhat a skill adds is a folder: supporting files and scripts next to the instructions, more frontmatter fields, and Claude loading the skill by itself when your request matches its description. A bare command file cannot have supporting files. A good rule: keep a one-file prompt as a command, and move it to a skill the day you want a bundled script or automatic loading. The details are in [Claude Code skills](\u002Fblog\u002Fclaude-code-skills).\n\nIf you do not want Claude to run a command on its own initiative, add `disable-model-invocation: true` to its frontmatter. That is what you want for anything with side effects, like deploys.\n\n## Where do named arguments and shell output fit?\n\nBoth come from the skill format, and since commands and skills are now one mechanism, the safest place to use them is a `SKILL.md`. Positional access is `$0`, `$1` (zero-based), or you declare names:\n\n```md\n---\ndescription: Migrate a component between languages\narguments: [component, from, to]\n---\n\nMigrate the $component component from $from to $to.\n```\n\n`\u002Fmigrate-component SearchBar JavaScript TypeScript` fills all three. Quote multi-word values. A line like !`git diff HEAD` runs the shell command before Claude sees the prompt and pastes its output in, which is how you give a command fresh context without asking Claude to run it.\n\n## A short list of commands people write\n\n- `\u002Fstandup`: summarize yesterday's commits across your branches.\n- `\u002Fchangelog`: draft release notes from the commits since the last tag.\n- `\u002Fpr-description`: write a pull request body from the branch diff.\n- `\u002Ftest-plan`: list the cases a change needs, without writing code.\n- `\u002Fexplain`: walk through a file for someone new to it.\n\nPick the one prompt you retype most this week and turn it into a file. For the habits around prompts themselves, see [how to prompt Claude Code](\u002Fblog\u002Fclaude-code-prompting-tips).\n\n## In Relais\n\nRelais' composer takes `\u002F` for commands and skills like the terminal does, and `@` to mention a file. Sessions are plain Claude Code sessions. Relais also has its own session templates, Markdown files in `.relais\u002Ftemplates\u002F` that set standing rules, skills, model and permissions for a session. See [features](\u002Ffeatures).","\u003Cp>Slash commands are what you type at the start of a Claude Code message to control the session: \u003Ccode>\u002Fclear\u003C\u002Fcode> empties the context, \u003Ccode>\u002Fcompact\u003C\u002Fcode> summarizes it, \u003Ccode>\u002Fpermissions\u003C\u002Fcode> edits your rules, \u003Ccode>\u002Finit\u003C\u002Fcode> writes a CLAUDE.md. Type \u003Ccode>\u002F\u003C\u002Fcode> alone to see everything available to you. You can also write your own as a Markdown file, though in current Claude Code a custom command is really a skill.\u003C\u002Fp>\n\u003Ch2 id=\"how-do-slash-commands-work\">\u003Ca class=\"anchor\" href=\"#how-do-slash-commands-work\" aria-label=\"Link to this section\">#\u003C\u002Fa>How do slash commands work?\u003C\u002Fh2>\n\u003Cp>A command is only recognized at the start of your message; whatever follows the name becomes its arguments. Type \u003Ccode>\u002F\u003C\u002Fcode> and the menu filters as you type. If you send a command while Claude is still answering, it is queued and runs after the current turn, except for a few that run immediately without interrupting (\u003Ccode>\u002Fstatus\u003C\u002Fcode>, \u003Ccode>\u002Ftasks\u003C\u002Fcode> and \u003Ccode>\u002Fusage\u003C\u002Fcode> among them).\u003C\u002Fp>\n\u003Cp>The menu mixes three kinds of entries: built-in commands, bundled skills that ship with Claude Code (such as \u003Ccode>\u002Floop\u003C\u002Fcode>, \u003Ccode>\u002Fsimplify\u003C\u002Fcode>, \u003Ccode>\u002Fdebug\u003C\u002Fcode>, \u003Ccode>\u002Fbatch\u003C\u002Fcode>), and your own commands and skills. You can even chain several skills in one message, for example \u003Ccode>\u002Fskill-a \u002Fskill-b do XYZ\u003C\u002Fcode>, which loads both and passes the trailing text to each.\u003C\u002Fp>\n\u003Cp>The exact list changes between versions, so treat \u003Ccode>\u002Fhelp\u003C\u002Fcode> and the \u003Ca href=\"https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002Fcommands\">commands reference\u003C\u002Fa> as the source of truth. These are the ones most people use.\u003C\u002Fp>\n\u003Ch2 id=\"which-built-in-commands-are-worth-learning\">\u003Ca class=\"anchor\" href=\"#which-built-in-commands-are-worth-learning\" aria-label=\"Link to this section\">#\u003C\u002Fa>Which built-in commands are worth learning?\u003C\u002Fh2>\n\u003Cp>\u003Cstrong>Managing the conversation\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Cdiv class=\"table-wrap\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Command\u003C\u002Fth>\u003Cth>What it does\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fclear\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>New conversation with empty context. Pass a name to label the old one in the \u003Ccode>\u002Fresume\u003C\u002Fcode> picker.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fcompact\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Free up context while continuing the same conversation, by summarizing it.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fresume\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Reopen a previous conversation.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Frewind\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Roll the conversation and\u002For the code back to an earlier point (aliases \u003Ccode>\u002Fcheckpoint\u003C\u002Fcode>, \u003Ccode>\u002Fundo\u003C\u002Fcode>).\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fcontext\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Show how full the context window is, as a colored grid, with suggestions.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fbtw\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Ask a side question without adding it to the conversation.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Cp>The context ones matter most. They are the difference between a session that stays sharp and one that forgets what you said an hour ago; the reasons are in \u003Ca href=\"\u002Fblog\u002Fclaude-code-context-window\">Claude Code&#39;s context window\u003C\u002Fa>.\u003C\u002Fp>\n\u003Cp>\u003Cstrong>Setting up the project\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Cdiv class=\"table-wrap\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Command\u003C\u002Fth>\u003Cth>What it does\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>\u003Ccode>\u002Finit\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Generate a starting \u003Ccode>CLAUDE.md\u003C\u002Fcode>. See \u003Ca href=\"\u002Fblog\u002Fclaude-md-guide\">how to write one\u003C\u002Fa>.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fmemory\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Edit CLAUDE.md files and turn auto memory on or off.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fpermissions\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Manage allow, ask and deny rules in an interactive dialog.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fmodel\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Switch model and save it as the default for new sessions.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fconfig\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Open settings, or set a value directly, like \u003Ccode>\u002Fconfig verbose=true\u003C\u002Fcode>.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fadd-dir &lt;path&gt;\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Give the session access to another directory.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Cp>\u003Cstrong>Extending and checking\u003C\u002Fstrong>\u003C\u002Fp>\n\u003Cdiv class=\"table-wrap\">\u003Ctable>\u003Cthead>\u003Ctr>\u003Cth>Command\u003C\u002Fth>\u003Cth>What it does\u003C\u002Fth>\u003C\u002Ftr>\u003C\u002Fthead>\u003Ctbody>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fskills\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>List available skills; you can filter by name, description or source.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fhooks\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>View hook configurations.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fmcp\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Manage MCP server connections.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fagents\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Create or manage subagents.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fplan [description]\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Enter \u003Ca href=\"\u002Fblog\u002Fclaude-code-plan-mode\">plan mode\u003C\u002Fa>, optionally with a first task.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Freview\u003C\u002Fcode>, \u003Ccode>\u002Fsecurity-review\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Review the current diff, or the branch&#39;s changes for security issues.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fdoctor\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Run a setup checkup; \u003Ccode>\u002Fdoctor prompt-audit\u003C\u002Fcode> audits your instruction files.\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fusage\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Show plan usage (\u003Ccode>\u002Fcost\u003C\u002Fcode> is an alias).\u003C\u002Ftd>\u003C\u002Ftr>\u003Ctr>\u003Ctd>\u003Ccode>\u002Fhelp\u003C\u002Fcode>, \u003Ccode>\u002Fpowerup\u003C\u002Fcode>\u003C\u002Ftd>\u003Ctd>Help, and short interactive lessons.\u003C\u002Ftd>\u003C\u002Ftr>\u003C\u002Ftbody>\u003C\u002Ftable>\u003C\u002Fdiv>\n\u003Ch2 id=\"how-do-you-write-your-own-slash-command\">\u003Ca class=\"anchor\" href=\"#how-do-you-write-your-own-slash-command\" aria-label=\"Link to this section\">#\u003C\u002Fa>How do you write your own slash command?\u003C\u002Fh2>\n\u003Cp>Create a Markdown file. The file name becomes the command name. A project command goes in \u003Ccode>.claude\u002Fcommands\u002F\u003C\u002Fcode> and is shared with your team through git. This one is \u003Ccode>.claude\u002Fcommands\u002Ffix-issue.md\u003C\u002Fcode>:\u003C\u002Fp>\n\u003Cpre class=\"code\" data-lang=\"md\">\u003Ccode>---\ndescription: Fix a GitHub issue end to end\nargument-hint: [issue-number]\n---\n\nFix GitHub issue $ARGUMENTS.\n\n1. Read the issue with `gh issue view $ARGUMENTS`.\n2. Find the code responsible and explain the cause in two sentences.\n3. Make the smallest fix and add a failing-then-passing test.\n4. Run `npm test` and `npm run lint`; fix what you broke.\n5. Summarize what changed. Do not commit.\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>Now \u003Ccode>\u002Ffix-issue 482\u003C\u002Fcode> expands the file with \u003Ccode>482\u003C\u002Fcode> in place of \u003Ccode>$ARGUMENTS\u003C\u002Fcode>. A few rules of thumb:\u003C\u002Fp>\n\u003Cul>\n\u003Cli>Write it as a prompt you would be happy to type, including the steps, the checks and what &quot;done&quot; means.\u003C\u002Fli>\n\u003Cli>Say what not to do (here, commit), so a command you run often has no side effects you did not intend.\u003C\u002Fli>\n\u003Cli>\u003Ccode>$ARGUMENTS\u003C\u002Fcode> is everything typed after the name. If you use no placeholder at all, Claude Code appends the arguments to the end for you.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Ch2 id=\"is-a-custom-command-different-from-a-skill\">\u003Ca class=\"anchor\" href=\"#is-a-custom-command-different-from-a-skill\" aria-label=\"Link to this section\">#\u003C\u002Fa>Is a custom command different from a skill?\u003C\u002Fh2>\n\u003Cp>Today they are the same thing. The official docs state that custom commands have been merged into skills: \u003Ccode>.claude\u002Fcommands\u002Fdeploy.md\u003C\u002Fcode> and \u003Ccode>.claude\u002Fskills\u002Fdeploy\u002FSKILL.md\u003C\u002Fcode> both create \u003Ccode>\u002Fdeploy\u003C\u002Fcode>, existing command files keep working, and if both exist with the same name the skill wins.\u003C\u002Fp>\n\u003Cp>What a skill adds is a folder: supporting files and scripts next to the instructions, more frontmatter fields, and Claude loading the skill by itself when your request matches its description. A bare command file cannot have supporting files. A good rule: keep a one-file prompt as a command, and move it to a skill the day you want a bundled script or automatic loading. The details are in \u003Ca href=\"\u002Fblog\u002Fclaude-code-skills\">Claude Code skills\u003C\u002Fa>.\u003C\u002Fp>\n\u003Cp>If you do not want Claude to run a command on its own initiative, add \u003Ccode>disable-model-invocation: true\u003C\u002Fcode> to its frontmatter. That is what you want for anything with side effects, like deploys.\u003C\u002Fp>\n\u003Ch2 id=\"where-do-named-arguments-and-shell-output-fit\">\u003Ca class=\"anchor\" href=\"#where-do-named-arguments-and-shell-output-fit\" aria-label=\"Link to this section\">#\u003C\u002Fa>Where do named arguments and shell output fit?\u003C\u002Fh2>\n\u003Cp>Both come from the skill format, and since commands and skills are now one mechanism, the safest place to use them is a \u003Ccode>SKILL.md\u003C\u002Fcode>. Positional access is \u003Ccode>$0\u003C\u002Fcode>, \u003Ccode>$1\u003C\u002Fcode> (zero-based), or you declare names:\u003C\u002Fp>\n\u003Cpre class=\"code\" data-lang=\"md\">\u003Ccode>---\ndescription: Migrate a component between languages\narguments: [component, from, to]\n---\n\nMigrate the $component component from $from to $to.\u003C\u002Fcode>\u003C\u002Fpre>\n\u003Cp>\u003Ccode>\u002Fmigrate-component SearchBar JavaScript TypeScript\u003C\u002Fcode> fills all three. Quote multi-word values. A line like !\u003Ccode>git diff HEAD\u003C\u002Fcode> runs the shell command before Claude sees the prompt and pastes its output in, which is how you give a command fresh context without asking Claude to run it.\u003C\u002Fp>\n\u003Ch2 id=\"a-short-list-of-commands-people-write\">\u003Ca class=\"anchor\" href=\"#a-short-list-of-commands-people-write\" aria-label=\"Link to this section\">#\u003C\u002Fa>A short list of commands people write\u003C\u002Fh2>\n\u003Cul>\n\u003Cli>\u003Ccode>\u002Fstandup\u003C\u002Fcode>: summarize yesterday&#39;s commits across your branches.\u003C\u002Fli>\n\u003Cli>\u003Ccode>\u002Fchangelog\u003C\u002Fcode>: draft release notes from the commits since the last tag.\u003C\u002Fli>\n\u003Cli>\u003Ccode>\u002Fpr-description\u003C\u002Fcode>: write a pull request body from the branch diff.\u003C\u002Fli>\n\u003Cli>\u003Ccode>\u002Ftest-plan\u003C\u002Fcode>: list the cases a change needs, without writing code.\u003C\u002Fli>\n\u003Cli>\u003Ccode>\u002Fexplain\u003C\u002Fcode>: walk through a file for someone new to it.\u003C\u002Fli>\n\u003C\u002Ful>\n\u003Cp>Pick the one prompt you retype most this week and turn it into a file. For the habits around prompts themselves, see \u003Ca href=\"\u002Fblog\u002Fclaude-code-prompting-tips\">how to prompt Claude Code\u003C\u002Fa>.\u003C\u002Fp>\n\u003Ch2 id=\"in-relais\">\u003Ca class=\"anchor\" href=\"#in-relais\" aria-label=\"Link to this section\">#\u003C\u002Fa>In Relais\u003C\u002Fh2>\n\u003Cp>Relais&#39; composer takes \u003Ccode>\u002F\u003C\u002Fcode> for commands and skills like the terminal does, and \u003Ccode>@\u003C\u002Fcode> to mention a file. Sessions are plain Claude Code sessions. Relais also has its own session templates, Markdown files in \u003Ccode>.relais\u002Ftemplates\u002F\u003C\u002Fcode> that set standing rules, skills, model and permissions for a session. See \u003Ca href=\"\u002Ffeatures\">features\u003C\u002Fa>.\u003C\u002Fp>\n",[15,18,21,24,27,30,33],{"id":16,"text":17},"how-do-slash-commands-work","How do slash commands work?",{"id":19,"text":20},"which-built-in-commands-are-worth-learning","Which built-in commands are worth learning?",{"id":22,"text":23},"how-do-you-write-your-own-slash-command","How do you write your own slash command?",{"id":25,"text":26},"is-a-custom-command-different-from-a-skill","Is a custom command different from a skill?",{"id":28,"text":29},"where-do-named-arguments-and-shell-output-fit","Where do named arguments and shell output fit?",{"id":31,"text":32},"a-short-list-of-commands-people-write","A short list of commands people write",{"id":34,"text":35},"in-relais","In Relais",[37,42,48],{"slug":38,"title":39,"description":40,"category":7,"categoryId":8,"keyword":41,"date":10,"minutes":11},"claude-code-beginners-guide","Claude Code for beginners: install, first session, habits","Install Claude Code, run your first session and learn the habits that matter: specific prompts, plan before edit, CLAUDE.md, small commits, a clean context.","claude code for beginners",{"slug":43,"title":44,"description":45,"category":7,"categoryId":8,"keyword":46,"date":10,"minutes":47},"claude-code-hooks","Claude Code hooks: run a script on every tool call","Claude Code hooks run your own script before or after a tool call. Learn the config format, exit codes and three working examples you can paste today.","claude code hooks",6,{"slug":49,"title":50,"description":51,"category":7,"categoryId":8,"keyword":52,"date":10,"minutes":11},"claude-code-permission-modes","Claude Code permission modes explained (and when to use each)","Default, acceptEdits, plan, auto, dontAsk and bypassPermissions: what each Claude Code permission mode allows, how to switch, and which one fits which job.","claude code permission modes",1794931098788]