Dashboard CLI

Command-line interface for the OpenClaw Dashboard API. Provides ergonomic commands for task management, research coordination, pipeline workflows, activity logging, briefings, insights, and cost tracking.

Binary: dashboard (installed globally via npm link)

Dashboard URL: Defaults to http://localhost:3000 (configurable via DASHBOARD_URL env var)

Quick Reference

Core Workflows

Morning Briefing:

bash

dashboard briefing

Task Workflow:

bash

# List next priority task
dashboard task list --status=backlog --priority=high --format=ids | head -1

# Start task
dashboard task update HEKTOR-003 --status=in-progress --assignee=hektor
dashboard activity log --type=task --agent=hektor --title="Started: HEKTOR-003"

# Complete task
dashboard task update HEKTOR-003 --status=done --actual-hours=3
dashboard activity log --type=task --agent=hektor --title="Completed: HEKTOR-003"

Research Workflow (Scout):

bash

# Check queue
dashboard research list --status=pending --assignee=scout

# Start research
dashboard research update RES-001 --status=in-progress
dashboard activity log --type=research --agent=scout --title="Started: RES-001"

# Complete research
dashboard research complete RES-001 --findings-file=findings.md
dashboard activity log --type=research --agent=scout --title="Completed: RES-001"

Commands

Tasks

List tasks:

bash

dashboard task list [--status=STATUS] [--priority=PRIORITY] [--project=PROJECT] [--assignee=ASSIGNEE] [--format=FORMAT]

Formats: table (default), json, yaml, ids

Get task details:

bash

dashboard task get <taskId> [--format=FORMAT]

Create task:

bash

dashboard task create --title="Task title" --project=PROJECT [--status=STATUS] [--priority=PRIORITY] [--assignee=ASSIGNEE] [--description=TEXT] [--deadline=ISO-DATE] [--estimated-hours=NUMBER]

Update task:

bash

dashboard task update <taskId> [--status=STATUS] [--priority=PRIORITY] [--assignee=ASSIGNEE] [--deadline=ISO-DATE] [--actual-hours=NUMBER] [--blocked-by=TASK_ID] [--blocks=TASK_ID]

Delete task:

bash

dashboard task delete <taskId>

Research

List research requests:

bash

dashboard research list [--status=STATUS] [--priority=PRIORITY] [--assignee=ASSIGNEE] [--format=FORMAT]

Get research details:

bash

dashboard research get <researchId> [--format=FORMAT]

Create research request:

bash

dashboard research create --question="Question text" [--context=TEXT] [--priority=PRIORITY] [--assignee=ASSIGNEE] [--deadline=ISO-DATE]

Update research:

bash

dashboard research update <researchId> [--status=STATUS] [--priority=PRIORITY] [--findings=TEXT] [--findings-file=PATH] [--linked-docs=PATH]

Complete research (shortcut):

bash

dashboard research complete <researchId> --findings=TEXT|--findings-file=PATH

Pipeline

List pipeline items:

bash

dashboard pipeline list [--type=TYPE] [--status=STATUS] [--project=PROJECT] [--assignee=ASSIGNEE] [--format=FORMAT]

Types: lead, competitor, opportunity
Status: research, qualified, approved, delivered

Get pipeline item:

bash

dashboard pipeline get <pipelineId> [--format=FORMAT]

Create pipeline item:

bash

dashboard pipeline create --name="Item name" --type=TYPE --project=PROJECT [--status=STATUS] [--assignee=ASSIGNEE] [--data=JSON]

Update pipeline item:

bash

dashboard pipeline update <pipelineId> [--status=STATUS] [--assignee=ASSIGNEE] [--data=JSON] [--name=NAME]

Delete pipeline item:

bash

dashboard pipeline delete <pipelineId>

Activity

Log activity:

bash

dashboard activity log --type=TYPE --agent=AGENT --title="Title text" [--details=TEXT|JSON] [--details-file=PATH] [--project=PROJECT]

Types: task, research, system, model, error, blocked, external, cost, pipeline, briefing

List activity:

bash

dashboard activity list [--type=TYPE] [--agent=AGENT] [--project=PROJECT] [--limit=NUMBER] [--format=FORMAT]

Projects

List projects:

bash

dashboard project list [--status=STATUS] [--all] [--format=FORMAT]

Get project:

bash

dashboard project get <projectId> [--format=FORMAT]

Create project:

bash

dashboard project create --name="Project name" [--emoji=EMOJI] [--color=HEX] [--charter=TEXT] [--charter-file=PATH]

Update project:

bash

dashboard project update <projectId> [--name=NAME] [--status=STATUS] [--charter=TEXT] [--charter-file=PATH]

Briefing

Get aggregated dashboard snapshot (open tasks, blockers, activity, pipeline, research):

bash

dashboard briefing [--format=FORMAT]

Output includes:

•Open tasks (by priority)
•In-progress tasks
•Blockers (>24h stuck)
•Recent activity (last 24h)
•Active projects
•Pipeline items (by status)
•Research queue (by status)

Use for: Morning briefing, evening summary, status checks

Insights

Get business metrics (Lead Gen, CI, Research, Tasks):

bash

dashboard insights [--format=FORMAT]

Output includes:

•Lead Gen: researched/qualified/delivered counts
•Competitive Intelligence: competitor moves, active tracking
•Research Queue: pending/in-progress/completed counts
•Tasks: total/backlog/in-progress/done counts
•Recent activity count

Cost Tracking

Get model usage and cost data:

bash

dashboard cost [--from=YYYY-MM-DD] [--to=YYYY-MM-DD] [--format=FORMAT]

Output includes:

•Total cost
•Budget status (monthly budget, used amount, percentage)
•Model split (Haiku/Sonnet/Opus counts + costs)
•Agent split (Hektor/Scout costs)
•Daily costs (time series)

Global Options

All commands support:

•--help — Show command help
•--json — Force JSON output (overrides --format)
•--quiet — Suppress non-essential output
•--verbose — Show debug info (HTTP requests, etc.)

Output Formats

table (default for list commands):

code

ID            Title                          Status        Priority
────────────────────────────────────────────────────────────────────
HEKTOR-001    Review Model Routing Framework backlog       high
HEKTOR-002    Dashboard Workflow Integration done          high

yaml (default for get/briefing/insights/cost):

yaml

id: HEKTOR-002
title: Dashboard Workflow Integration
status: done
priority: high
assignee: hektor
projectId: hektor-setup

json:

json

{
  "id": "HEKTOR-002",
  "title": "Dashboard Workflow Integration",
  "status": "done",
  "priority": "high"
}

ids (for piping):

code

HEKTOR-001
HEKTOR-002

Integration Patterns

Hektor Morning Briefing (Cron: 07:00)

bash

# Get snapshot
dashboard briefing

# Check for blockers
BLOCKERS=$(dashboard task list --status=in-progress | grep "⚠️" | wc -l)
if [ "$BLOCKERS" -gt 0 ]; then
  echo "🚨 Blockers detected"
  # Escalate to #alerts
fi

# Pull next task
NEXT_TASK=$(dashboard task list --status=backlog --priority=high --format=ids | head -1)
if [ -n "$NEXT_TASK" ]; then
  dashboard task update "$NEXT_TASK" --status=in-progress --assignee=hektor
  dashboard activity log --type=task --agent=hektor --title="Started: $NEXT_TASK"
fi

Scout Research Check-In (Cron: 08:00)

bash

# Check pending queue
PENDING=$(dashboard research list --status=pending --assignee=scout --format=ids)
if [ -n "$PENDING" ]; then
  echo "📋 Scout has pending research requests"
  # Process next research
fi

Task Completion with Activity Logging

bash

TASK_ID="HEKTOR-003"

# Complete task
dashboard task update "$TASK_ID" --status=done --actual-hours=3

# Log completion
dashboard activity log \
  --type=task \
  --agent=hektor \
  --title="Completed: $TASK_ID" \
  --details='{"outcome":"Implementation successful","blockers":"none"}'

Model Switch Logging

bash

dashboard activity log \
  --type=model \
  --agent=hektor \
  --title="Model: Sonnet for HEKTOR-003" \
  --details='{"model":"sonnet","reason":"Complex architecture decisions","taskId":"HEKTOR-003","estimatedCost":0.0075}'

Workflow Tiers

Tier 1: Full Workflow (>30min tasks)

•Morning/Evening Briefing
•Task Start + Done + Activity Log
•Model routing logged
•Example: Large implementations, architecture work

Tier 2: Light Workflow (10-30min tasks)

•Task Done + Activity Log (completion only)
•Example: Config patches, skill installations, bug fixes

Tier 3: No Tracking (<10min routine)

•No dashboard interaction unless blocked
•Example: File operations, quick checks, routine maintenance

Error Handling

Exit codes:

•0 — Success
•1 — General error
•2 — API error
•3 — Invalid arguments

HTTP errors:

•404 → "Error: Resource not found"
•500 → "Error: API error: <message>"
•Network error → "Error: Cannot connect to Dashboard API at <url>"

Installation

bash

cd ~/openclaw-dashboard
npm link

Verify:

bash

dashboard --help

Resources

references/workflow-examples.md

See references/workflow-examples.md for complete agent workflow patterns including:

•Hektor daily operations (Morning/Evening)
•Scout research workflows
•Cross-agent coordination (Pipeline handoffs)
•Cost tracking integration
•Blocker detection and escalation