How this skill is triggered — by the user, by Claude, or both
Slash command
/crew:dashboardThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Display current workflow state, progress, and cost/token analytics.
Display current workflow state, progress, and cost/token analytics.
Invoked as: $ARGUMENTS
cat .caw/task_plan.md 2>/dev/null | head -20 || echo "(no task plan)"cat .caw/metrics.json 2>/dev/null | head -10 || echo "(no metrics yet)"cat .caw/mode.json 2>/dev/null || echo "(default mode)"/crew:dashboard # Standard status
/crew:dashboard --verbose # Detailed with file lists
/crew:dashboard --cost # Cost breakdown by model tier
/crew:dashboard --tokens # Token usage analysis
/crew:dashboard --sessions # Multi-session comparison
/crew:dashboard --worktrees # Show active worktree status
/crew:dashboard --agents # Show background agent status
/crew:dashboard --export # Export metrics to JSON
/crew:dashboard --all # Everything
| Flag | Description |
|---|---|
--verbose | Context files and recent activity |
--cost | Cost breakdown: current, weekly, monthly, top drivers |
--tokens | Token analysis: I/O ratio, category breakdown, tips |
--sessions | Compare last 5 sessions: tokens, cost, steps |
--worktrees | Active worktree status and progress |
--agents | Background agent status and duration |
--export | Export to .caw/analytics_export_[date].json |
--all | All of the above |
Look for .caw/task_plan.md. Show help if not found.
progress_percent = (completed / total) * 100
progress_bar = "█" * filled + "░" * empty
Read .caw/mode.json for DEEP_WORK or NORMAL mode.
Standard Output:
Workflow Status
Task: [Title]
Status: [status]
Mode: [DEEP WORK | NORMAL]
Phase [N]: [Phase Name]
|- [N.1] [Step] Complete
|- [N.2] [Step] In Progress <- current
|- [N.3] [Step] Pending
Progress: [X]% ([completed]/[total])
WORKFLOW ANALYTICS
Session: abc123 | Duration: 1h 30m
TOKEN USAGE
Input: 45,000 (79%) | Output: 12,000 (21%) | Total: 57,000
COST BREAKDOWN
Haiku: 15K/$0.02 (4%) | Sonnet: 35K/$0.15 (29%) | Opus: 7K/$0.35 (67%)
TOTAL: $0.52
OPTIMIZATION INSIGHTS
Opus 13% of tokens drove 67% of cost
Eco mode would save ~$0.18 (35%)
| Icon | Status | Meaning |
|---|---|---|
| Done | Complete | Finished |
| Working | In Progress | Working |
| Pending | Pending | Not started |
| Blocked | Blocked | Cannot proceed |
| Skipped | Skipped | Bypassed |
Shows phase-based and step-based worktrees with branch, directory, status, and progress.
Shows background agents with task ID, step, status, duration.
Will:
.caw/task_plan.md, .caw/metrics.json, .caw/sessions/*.json, .caw/mode.json.caw/analytics_export_*.json (with --export)dependency-analyzer for parallel opportunitiesWon't:
npx claudepluginhub jaebit/claudemate --plugin crewShows a single-screen work status dashboard with task state, epic progress, knowledge flywheel health, and recent activity. Useful for getting oriented or deciding next actions.
Displays Plan-Build-Run project status dashboard, progress, blockers from .planning files, and suggests next actions.
Reconstructs AI agent session history to summarize project state, decisions, and workstreams when onboarding to a brownfield repo. Recommends gates from open decisions.