Core Commands
The command set every Claude Code session should start from.
Key takeaways
- Covers the everyday command set: session start (
/help,/status,/doctor), continue-or-reset, model and usage controls, and pre-ship review. /usageis the single cost-and-quota command;/costand/statsare aliases, with/statsopening on the Stats tab.- On the Anthropic API
/modelmapsfableto Fable 5,opusto Opus 5, andsonnetto Sonnet 5;bestprefers Fable when available. Opus 5 requires Claude Code v2.1.219+ and has native 1M context. - After 2026-07-07, subscription-plan access to Fable 5 requires usage credits; API integrations must treat HTTP 200 responses with
stop_reason: "refusal"as a separate successful response state. /rewindrecovers a previous state but is not a substitute for git;/branchswitches to a new branch,/forkstarts a copied background session, and/subtaskdelegates within this conversation./pr-comments(v2.1.91) and/vim(v2.1.92) were removed; view PR comments by asking Claude, and toggle Vim mode via/config.
Session Start
| Command | Purpose | Team habit |
|---|---|---|
/help | Show help and the commands available in the current session. | Use it as the final source of truth because availability varies by platform, plan, and environment. |
/status | Open the Settings interface on the Status tab, showing version, model, account, and connectivity. Works while Claude is responding. | Capture it when debugging environment drift. |
/config | Open the Settings interface to adjust theme, model, output style, and other preferences. Alias: /settings. | Remember that v2.1.119+ persists settings to ~/.claude/settings.json. |
/doctor | Run a full setup checkup that diagnoses and can fix issues. /checkup is its alias (v2.1.205+). | Run it before changing project config when a teammate reports a broken setup. |
Continue or Reset
| Command | Purpose | Use carefully |
|---|---|---|
/resume | Resume a conversation by ID or name, or open the session picker. As of v2.1.144 background sessions appear in the picker marked bg. Alias: /continue. | Large stale sessions may be summarized before reload. |
/clear | Start a new conversation with empty context; project memory is kept. Aliases: /reset, /new. | The previous conversation stays available through /resume. |
/compact | Free up context by summarizing the conversation while continuing the same thread. | Add focus instructions when the task has a clear next step. |
/rewind | Rewind the conversation and/or code to a previous point, or summarize from a selected message. Aliases: /checkpoint, /undo. | Treat it as a recovery tool, not a replacement for git discipline. |
/branch | Create a branch of the current conversation and switch into it. | Name branches with /rename; the original remains reachable via /resume. |
/fork [prompt] | Copy the current conversation into a separate background session while the original continues (v2.1.212+). | Use it for an independently resumable side path. |
/subtask <prompt> | Delegate inherited-context work inside the current conversation and return the result here (v2.1.212+). | Use it for bounded research or implementation, not an independent session. |
/exit | Leave the CLI. Alias: /quit. In an attached background session this detaches and the session keeps running. | Make handoff notes before closing long-running work. |
Model and Usage
| Command | Purpose | Operating note |
|---|---|---|
/model | Switch the model and save it as your default; press s in the picker to switch for the current session only. | On Anthropic API, fable maps to Fable 5, opus to Opus 5, and sonnet to Sonnet 5. Pin full IDs across providers. |
/advisor [model|off] | Consult a second model for the current task. | fable is available as an advisor option when your organization has Fable 5 access. |
/effort | Set the model effort level: low, medium, high, xhigh, max, or ultracode; auto resets to the model default. | Available levels depend on the model; max and ultracode are session-only. Fable 5 and Opus 4.8 default to high effort, so reserve xhigh for the hardest tasks. |
/fast | Toggle fast mode on or off. | Current support covers Opus 5 and Opus 4.8; Opus 4.7 was removed in v2.1.219. |
/usage | Show session cost, plan usage limits, and activity stats. On paid plans it breaks usage down by skill, subagent, plugin, and MCP server. | Standardize on this command; /cost and /stats are aliases (/stats opens on the Stats tab). |
/usage-credits | Configure usage credits to keep working when you hit a plan limit. Previously /extra-usage. | Confirm billing and admin policy before making it a default workflow. |
/goal | Set a completion condition; Claude keeps working across turns until it is met. Clear it with clear, stop, off, reset, none, or cancel. | Use it only when the success condition is concrete. |
Deprecation warnings became more explicit in v2.1.187. If /model, /status, or startup output warns about an older full model ID, update team settings before treating the session as reproducible.
Fable 5 Plan Access
The temporary subscription allowance ended on 2026-07-07. As of 2026-07-19, keep seat-based subscription usage separate from API consumption:
| Access path | Current rule | Operating note |
|---|---|---|
| Pro, Max, and Team | Usage credits required | Fable 5 is unavailable when credits are disabled or the balance is exhausted. |
| Standard Enterprise seat | Usage credits required | Standard seats were not part of the temporary included allowance. |
| Premium/select Enterprise seat | Usage credits required | Up to 50% of seat usage was included only through 2026-07-07. |
| Claude API and consumption-based Enterprise | Token-based API billing | Separate from weekly seat allowances; standard Fable 5 API pricing applies. |
Usage credits are billed separately from the subscription at standard API rates. Configure payment,
prepaid balance, monthly caps, and auto-reload in Settings > Usage; use /usage-credits from Claude
Code to open the corresponding account flow. Teams should set an admin policy and spending cap before
making Fable the default path.
Fable 5 Refusal and Fallback
A Fable safety-classifier refusal is a successful HTTP 200 Messages API response. Branch on
stop_reason, not on HTTP errors, empty content, or stop_details:
{
"model": "claude-fable-5",
"content": [],
"stop_reason": "refusal",
"stop_details": {
"type": "refusal",
"category": "cyber",
"explanation": "..."
}
}stop_details.category may identify an area such as cyber or bio, but it can be null and the
human-readable explanation is not a stable parsing contract.
| Fallback path | Availability | Implementation rule |
|---|---|---|
| Server-side | Beta on the Claude API and Claude Platform on AWS | Pass up to three models in fallbacks with the server-side-fallback-2026-06-01 beta header. |
| SDK middleware | Supported platforms | Reuse conversation fallback state so follow-up turns stay pinned to the model that accepted the request. |
| Manual retry | Any platform or language | Detect the refusal, retry on another model, and redeem the fallback-credit-2026-06-01 credit token. |
- Do not combine server-side
fallbacksand the SDK refusal-fallback middleware on one request. - Message Batches can return refused items as
result.type: "succeeded"; inspectstop_reason. Server-sidefallbacksis not supported for batches. - Claude Code
fallbackModeland--fallback-modelhandle availability failures. They are not the same setting as the Messages API refusal fallback. - A request refused before output is not billed. Server-side fallback and the SDK middleware apply fallback credit automatically; manual retry logic must redeem the credit token to avoid paying the prompt-cache write cost twice when switching models.
- Instrument refusals and fallback-served responses separately. HTTP-error monitoring alone cannot reveal this path.
Prompts, skills, and harness instructions that ask Fable to echo, transcribe, or expose its internal
reasoning can trigger the reasoning_extraction refusal category and increase fallback to Opus 4.8.
Ask for a concise rationale, verification evidence, and user-facing progress instead. When an API
integration needs reasoning visibility, consume the structured adaptive-thinking blocks.
Setup and Review
A few commands are worth knowing before you ship a change or start a new repo.
| Command | Purpose | Operating note |
|---|---|---|
/init | Initialize the project with a CLAUDE.md guide. | When a CLAUDE.md already exists it suggests improvements rather than overwriting it. Set CLAUDE_CODE_NEW_INIT=1 for an interactive flow that also walks through skills, hooks, and personal memory files. |
/plan | Enter plan mode directly from the prompt; pass a description to start immediately. | Use it before a large change to capture the approach and risks before editing. |
/code-review | Review the current diff for correctness bugs and cleanups; pass --fix to apply findings or --comment to post inline GitHub PR comments. | User-invoked only since v2.1.215; runs in a background subagent since v2.1.218. |
/review [PR] | Run a fast single-pass PR review in the current session. | Since v2.1.202, use /code-review <level> <PR#> when you want a multi-agent review at a chosen effort level. |
/fewer-permission-prompts | Scan your transcripts for common read-only Bash and MCP calls and add a prioritized allowlist to project .claude/settings.json. | Run it once the project's safe-tool patterns have stabilized. |
Removed commands
/pr-comments was removed in v2.1.91 (ask Claude directly to view PR comments instead), and /vim was removed in v2.1.92 (toggle Vim editing via /config → Editor mode).
Practical Startup Routine
/status
/doctor
/model
/usageFor a long task, add:
/plan describe the change and risks before editingOfficial Fable References
- Introduction and API differences: https://platform.claude.com/docs/en/about-claude/models/introducing-claude-fable-5-and-claude-mythos-5
- Refusals and fallback: https://platform.claude.com/docs/en/build-with-claude/refusals-and-fallback
- Fallback credit: https://platform.claude.com/docs/en/build-with-claude/fallback-credit
- Prompting Claude Fable 5: https://platform.claude.com/docs/en/build-with-claude/prompt-engineering/prompting-claude-fable-5
- Usage credits: https://support.claude.com/en/articles/12429409-manage-usage-credits-for-paid-claude-plans