Mission Control V1 — OpenClaw Skill
A complete operations dashboard for managing OpenClaw agent fleets. This is a self-contained skill package — it includes the full dashboard application, lifecycle hook, agent instructions, and all documentation.
One command to install everything:
./setup.sh
What's Inside
mission-control-skill/
├── SKILL.md ← Agent instructions (injected into system prompt)
├── README.md ← This file
├── setup.sh ← All-in-one installer
├── hook.ts ← Lifecycle hook (agents → dashboard bridge)
├── docs/
│ ├── GETTING-STARTED.md ← Step-by-step beginner guide
│ ├── API-REFERENCE.md ← All 60+ API endpoints
│ ├── HOOK-EVENTS.md ← Every hook event with JSON examples
│ ├── LIBRARY-GUIDE.md ← Document Library usage
│ ├── TROUBLESHOOTING.md ← Common problems and fixes
│ ├── DOCKER.md ← Docker deployment guide
│ └── CHANGELOG.md ← Version history
└── app/
├── backend/ ← Express + SQLite (11 route files, 3 services)
│ ├── migrations/ ← 4 SQL migration files
│ ├── src/
│ │ ├── routes/ ← projects, tasks, agents, inbox, library, etc.
│ │ └── services/ ← SSE events, gateway sync, stall detector
│ └── .env.example
├── frontend/ ← React dashboard (8 views)
│ └── src/pages/ ← Dashboard, Inbox, Pipeline, Projects,
│ Library, Agents, Costs, Activity
├── package.json ← Root scripts (install:all, dev)
├── Dockerfile
└── docker-compose.yml
Installation
Option A: Interactive Setup
cd mission-control-skill
./setup.sh
The script asks for your gateway URL, generates a secret, installs the app, configures the hook, and tells you how to start.
Option B: Automated Setup
./setup.sh --auto
Uses all defaults: port 8000, auto-generated secret, local gateway at ws://127.0.0.1:18789.
Option C: Docker
./setup.sh --docker
Uses Docker Compose instead of local Node.js.
Option D: From ClawHub
clawhub install tsu-mission-control
cd ~/.openclaw/skills/mission-control
./setup.sh
Starting the Dashboard
cd ~/.openclaw/mission-control
npm run dev
Open http://localhost:4173 in your browser.
What the Skill Does for Agents
When this skill is loaded into an agent's context (via the SKILL.md file), the agent learns how to:
- Report task progress — start, progress, review, complete, fail
- Submit work for review — with deliverables, checklists, and summaries
- Publish documents — research, reports, and docs go to the Library
- Request approvals — workflow gates for risky actions
- Submit project requests — when agents identify new work
- Report costs — token usage per session
All communication happens via HTTP POST to the Mission Control backend. The lifecycle hook (hook.ts) handles the translation from OpenClaw events to API calls.
Dashboard Views
| View | What It Does |
|---|---|
| Dashboard | Stats, pipeline overview, agent status, alerts, recent activity |
| Inbox | Unified queue: requests + reviews + approvals in one sorted list |
| Pipeline | Kanban board with drag-and-drop and enforced stage transitions |
| Projects | Project CRUD with progress bars and agent assignment |
| Library | Document catalog with collections, search, multi-format reader |
| Agents | Agent monitoring with status, model info, task counts |
| Costs | Token spend tracking with daily trends and breakdowns |
| Activity | Chronological feed of all system events |
Documentation
| Doc | What It Covers |
|---|---|
| GETTING-STARTED.md | Step-by-step guide for first-time setup. Assumes no technical knowledge. |
| API-REFERENCE.md | Every API endpoint with methods, parameters, and example payloads. |
| HOOK-EVENTS.md | All 12 hook events agents can send, with complete JSON examples. |
| LIBRARY-GUIDE.md | How the Library works: collections, types, tags, formats, versioning, search. |
| TROUBLESHOOTING.md | Solutions for install errors, empty dashboards, hook failures, and more. |
| DOCKER.md | Running everything in Docker containers. |
| CHANGELOG.md | Version history and what changed. |
Publishing to ClawHub
clawhub publish ./mission-control-skill \
--slug tsu-mission-control \
--name "Mission Control" \
--version 1.0.0 \
--tags latest
License
MIT