The Command Centre for AI Agent Fleets
Monitor, manage, and optimize every agent in your fleet from a single pane of glass.
Cost analytics · Competitor intelligence · Memory health · Cron pipelines · Kanban workflows · Interactive chat.
100% local-first. Zero cloud dependency. Your data never leaves your machine.
Running AI agents at scale means juggling costs, context windows, memory drift, competitor landscapes, and coordination across dozens of autonomous processes. OpenClaw Dashboard gives you a single local-first command centre that reads directly from your OpenClaw installation -- no cloud dependency, no telemetry, full control.
New in v2.0 -- 10-step onboarding wizard, project workspaces with competitor scanning, market intelligence feeds, practitioner signal aggregation, visual cron pipeline builder, memory health AI analysis, and a complete Apple-inspired glass design system.
Real-time monitoring of multi-agent systems with constellation graph visualization, execution traces, context health bars, and drift detection. Four home views -- Org Map, Grid, Feed, and Constellation -- let you choose how to observe your fleet.
Discover, track, and profile competitors across your market. Add competitors manually or let AI discover them. Drill into per-competitor detail pages with SWOT analysis, update tracking, and category organization.
Per-agent and per-provider spend attribution with token-level granularity. Optimization scoring, cache savings estimation, anomaly detection, week-over-week trending, and per-model breakdowns.
Visual DAG editor for job dependencies powered by @xyflow/react. Track execution history with cost attribution, configure delivery rules, and build dependency chains between scheduled tasks.
Real-time SSE streaming chat with multimodal support. Upload images, review per-agent conversation history, and use slash commands (/help, /status, /cost) to query your fleet.
Full agile project management with columns (todo / in-progress / done / blocked), agent assignment, priority labels, ticket chat threads, and automation rules.
AI-driven analysis of agent memory with staleness detection, completeness checks, editing hints, and one-click reindex controls. Know when an agent's context is drifting before it causes problems.
Automated competitive intelligence feeds with signal categorization. Your agents scan the market and surface trends, competitor moves, and technology shifts.
Community signal aggregation from Reddit, forums, and social media. Surface pain points, feature requests, and positive signals from real practitioners in your domain.
Curated technology radar with category filtering. Stay on top of framework releases, AI model updates, and tooling changes relevant to your stack.
Organize your intelligence feeds, competitors, and reference files into named projects. Each project gets its own icon, description, and URL. Create, edit, and delete projects from the sidebar.
Full UI for accent color, portal branding (name, subtitle, logo), agent avatars, operator profile, and theme customization. 12 accent color presets + custom color picker.
| Feature | Description |
|---|---|
| 10-Step Onboarding Wizard | Guided setup with auto-detection, gateway testing, agent discovery, and feature selection |
| Global Search | Unified search across agents, crons, memory, costs, and all data types (Cmd+K) |
| Live Stream Widget | Real-time activity log panel accessible from any page |
| Agent Detail Pages | Per-agent view with SOUL.md personality viewer and conversation history |
| Activity Console | Browseable logs with filtering by source (cron, config, error) |
| Security Posture | Device auditing, key rotation reminders, configuration change tracking |
| Reference Files | Document storage with TipTap rich text editor, tags, and full-text search |
| Dynamic Favicon | Theme-aware favicon that switches between light/dark based on system preference |
| Voice Features | Text-to-speech (ElevenLabs) and speech-to-text transcription |
| Design System QA | Component-level fidelity scoring and token reuse tracking |
| Release Tracker | Version history and deployment notes |
| Docker Support | Production-ready multi-stage Dockerfile with docker-compose |
| View | Preview |
|---|---|
| Agent Fleet | |
| Competitor Intelligence | |
| Settings | |
| Cron Pipelines | |
| Activity Log |  |
| Security |  |
- Node.js 20+ (LTS recommended)
- npm 10+ or equivalent package manager
- OpenClaw installed locally (provides agent session data)
# Clone the repository
git clone https://github.com/ChristianAlmurr/openclaw-dashboard
cd openclaw-dashboard
# Install dependencies
npm install
# Configure environment
cp .env.local.example .env.local
# Edit .env.local with your paths (see Environment Variables below)
# Start the development server
npm run devOpen http://localhost:3333 in your browser. The 10-step onboarding wizard will guide you through initial configuration on first launch.
| Command | Description |
|---|---|
npm run dev |
Start development server on port 3333 |
npm run build |
Create production build |
npm start |
Start production server on port 3333 |
npm run lint |
Run ESLint |
npm test |
Run Vitest test suite |
docker compose up -d
docker compose logs -f mission-controldocker build -t openclaw-dashboard .
docker run -d \
--name openclaw-dashboard \
-p 3333:3333 \
-v ~/.openclaw-dashboard:/data/db \
-v ~/.openclaw:/data/openclaw:ro \
-e OPENCLAW_HOME=/data/openclaw \
-e DATA_DIR=/data/db \
openclaw-dashboardCopy .env.local.example to .env.local and configure:
| Variable | Default | Description |
|---|---|---|
OPENCLAW_HOME |
~/.openclaw |
Path to your OpenClaw data directory |
WORKSPACE_PATH |
~/.openclaw/workspace |
Path to your agent workspace |
OPENCLAW_BIN |
Auto-detected | Path to the OpenClaw CLI binary |
OPENCLAW_GATEWAY_TOKEN |
Auto-detected | Gateway authentication token |
OPENCLAW_GATEWAY_PORT |
18789 |
Gateway HTTP port |
PROJECT_REPO_PATH |
-- | Project repository path (enables task/sprint parsing) |
DATA_DIR |
~/.openclaw-dashboard |
SQLite database directory |
GITHUB_PAT |
-- | GitHub personal access token for private repos |
| Layer | Technologies |
|---|---|
| Frontend | Next.js 16, React 19, TypeScript 5.9, Tailwind CSS 4 |
| State | TanStack React Query v5, Zustand, React Context |
| UI | Radix UI, Lucide React, Recharts v3, @xyflow/react, TipTap |
| Backend | Next.js API Routes (52 endpoints), SSE streams |
| Database | SQLite via better-sqlite3 |
| Scheduling | node-cron, Chokidar file watchers |
| Integrations | OpenAI SDK, simple-git, ElevenLabs |
| Testing | Vitest, React Testing Library |
~/.openclaw/ Agent sessions, gateway logs, config
|
v
[Parsers & Watchers] Real-time log watchers + session parsers
|
v
[SQLite Database] Local persistence (zero cloud dependency)
|
v
[Next.js API Routes] 52 REST endpoints + SSE streams
|
v
[React Frontend] React Query -> Zustand -> Apple glass UI
On first launch, the 10-step wizard walks you through:
- Welcome -- Product overview
- Prerequisites -- Auto-detect Node.js, CLI, workspace, gateway
- Workspace -- Configure data directory
- Gateway -- Test connectivity
- Agent Discovery -- Scan and display your agents
- Features -- Toggle modules (fleet, chat, kanban, crons, memory, costs, etc.)
- Appearance -- Name, accent color, branding
- Notifications -- Cost alerts, key rotation reminders
- Integrations -- GitHub, optional webhooks
- Tour -- Feature walkthrough
Re-run anytime from Settings or /setup.
Apple-inspired glass aesthetic with full accent color customization. 12 presets + custom color picker. Design tokens defined as CSS custom properties in the Tailwind v4 @theme layer.
npm test # Run all tests
npx vitest # Watch mode
npx vitest --coverage # With coverageWe welcome contributions. See CONTRIBUTING.md for development setup, coding standards, and pull request guidelines.
MIT -- Copyright (c) 2026 OpenClaw Contributors