/setup

description: Configure claude-dashboard status line settings
argument-hint: "[displayMode] [language] [plan] | custom \"widgets\""
allowed-tools: Read, Write, Bash(node:*), AskUserQuestion

Claude Dashboard Setup

Configure the claude-dashboard status line plugin with widget system support.

Arguments

• **No arguments**: Interactive mode (asks questions)
• **With arguments**: Direct configuration mode

Direct Mode Arguments

• `$1`: Display mode
• `compact` (default): 1 line (model, context, cost, rateLimit5h, rateLimit7d, rateLimit7dSonnet, zaiUsage)
• `normal`: 2 lines (+ projectInfo, sessionId, sessionDuration, burnRate, todoProgress)
• `detailed`: 6 lines (+ sessionName, tokenSpeed, depletionTime, configCounts, toolActivity, agentStatus, cacheHit, promptCache, performance, tokenBreakdown, forecast, budget, todayCost, codexUsage, antigravityUsage, linesChanged, outputStyle, version, peakHours, lastPrompt, vimMode, apiDuration, tagStatus)
• `custom`: Custom widget configuration (requires `$4`)

• `$2`: Language preference
• `auto` (default): Detect from system language
• `en`: English
• `ko`: Korean

• `$3`: Subscription plan
• `max` (default): Shows 5h + 7d rate limits, plus model-scoped weekly limits (Fable)
• `ultra`: Same as `max`. Verified accounts have no 5h limit, so that widget hides itself
• `pro`: Shows 5h + 7d rate limits

• `$4`: Custom lines (only for `custom` mode)
• Format: `"widget1,widget2|widget3,widget4"`
• `|` separates lines
• Example: `"model,context,cost|projectInfo,todoProgress"`

Available Widgets

| Widget | Description | |--------|-------------| | `model` | Model name and version with emoji (e.g. `Opus 5.5`), effort level (Opus/Sonnet/Fable), fast mode (Opus) | | `context` | Progress bar, percentage, tokens | | `contextBar` | Progress bar only (sub-widget of `context`) | | `contextPercentage` | Percentage only (sub-widget of `context`) | | `contextUsage` | Token count only, e.g. `42K/200K` (sub-widget of `context`) | | `cost` | Session cost in USD | | `rateLimit5h` | 5-hour rate limit (hidden on verified Ultra accounts) | | `rateLimit7d` | 7-day rate limit (Pro/Max/Ultra) | | `rateLimit7dSonnet` | 7-day Sonnet limit (Max) — deprecated ~2026-06, merged into the unified weekly bucket at Sonnet 5 launch; stays hidden while the API returns null | | `rateLimit7dFable` | 7-day Fable limit (Max/Ultra) | | `projectInfo` | Directory name + git branch + ahead/behind (↑↓), subpath from project_dir, worktree indicator | | `configCounts` | CLAUDE.md, AGENTS.md, rules, MCPs, hooks, +Dirs counts | | `sessionId` | Session ID (short 8 chars) | | `sessionIdFull` | Session ID (full UUID) | | `sessionDuration` | Session duration | | `toolActivity` | Running/completed tools with targets (e.g., `Read(app.ts)`) | | `agentStatus` | Subagent progress | | `todoProgress` | Todo completion rate | | `burnRate` | Token consumption per minute, session average (input + cache write + output) | | `cacheHit` | Cache hit rate percentage (last request) | | `promptCache` | Session prompt-cache health: ♨️ warm with time left before it goes cold / ❄️ cold, hit ratio, miss count (Claude Code ≥ 2.1.251) | | `promptCacheState` | Warm/cold icon + time left only (sub-widget of `promptCache`) | | `promptCacheHit` | Session cache hit ratio only (sub-widget of `promptCache`) | | `promptCacheMisses` | Miss count only (sub-widget of `promptCache`) | | `depletionTime` | Estimated time to rate limit | | `codexUsage` | OpenAI Codex CLI usage (auto-hide if not installed) | | `geminiUsage` | Google Gemini CLI usage - current model (auto-hide if not installed or on a retired personal tier; not in `detailed`, opt in with `G`) | | `geminiUsageAll` | Google Gemini CLI usage - all models (auto-hide if not installed) | | `antigravityUsage` | Google Antigravity CLI - weekly quota by model family (auto-hide if not installed) | | `antigravityUsageAll` | Google Antigravity CLI - per-model quota (auto-hide if not installed) | | `zaiUsage` | z.ai/ZHIPU usage (auto-hide if not using z.ai) | | `tokenBreakdown` | Input/output/cache write/read token breakdown | | `performance` | Composite efficiency badge (cache hit + output ratio) | | `forecast` | Estimated hourly cost based on session rate | | `budget` | Daily spending vs configured budget limit (requires `dailyBudget` in config) | | `tokenSpeed` | Output token generation speed, session average (e.g., `67 tok/s`) | | `tokenSpeedLast` | Output token generation speed of the most recent response (e.g., `last 150 tok/s`) | | `sessionName` | Session name from /rename command | | `todayCost` | Total spending across all sessions today | | `linesChanged` | Uncommitted lines added/removed, including untracked files (+N -N) | | `outputStyle` | Current output style (hidden when "default") | | `version` | Claude Code version display | | `vimMode` | Vim mode (NORMAL/INSERT), auto-hides when vim disabled | | `apiDuration` | API time as % of total session time | | `peakHours` | Peak hours indicator with countdown (weekdays 5-11 AM PT) | | `tagStatus` | Commits ahead of matched git tags (uses `tagPatterns` config, default `["v*"]`) | | `slashCommand` | Active slash command for the current turn (🎯); cleared by next plain-text message | | `agentMode` | Session agent identity: 👤 custom agent (via `/agent <name>`) or 🤖 subagent type |

Tasks

1. Determine configuration

**If no arguments provided (interactive mode):**

Use AskUserQuestion to ask the user. Batch independent questions into a single AskUserQuestion call (max 4 per call) to minimize back-and-forth.

**Turn 1** — Ask all 4 questions in a single AskUserQuestion call: 1. Display mode — MUST include `markdown` field on each option for visual preview:

• compact (recommended), markdown:

     ◆ Opus 5.5(X) │ ██░░ 80% │ $1.25 │ 5h: 42% │ 7d: 69%

• normal, markdown:

     ◆ Opus 5.5(X) │ ██░░ 80% │ $1.25 │ 5h: 42% │ 7d: 69%
     📁 project (main ↑3) │ 🔑