Overview
The workspace overview is the first page you see after logging in. It answers: “Is everything OK, or do I need to act?”
Layout
Stat cards (top row)
Four summary cards:
| Card | Shows |
|---|
| Agents | Active / idle / terminated count |
| Uptime | Aggregate online rate across all agents |
| Traces | Tool calls recorded today |
| Blocked | Blocked operations count and percentage today |
Token usage by model
A horizontal bar chart showing token consumption over the last 7 days, broken down by LLM model (Claude Sonnet, GPT-4o, Gemini 2.5 Pro, etc.).
- OpenClaw agents: token data is captured by the plugin.
- MCP Gateway agents: requires the framework to report token counts via metadata headers (optional).
Alerts & suggestions
A feed of active risk alerts and AI-suggested actions:
- 🔴 Critical: high block rate, security anomalies.
- ⚠️ Warning: token spikes, latency spikes, offline agents.
- Each alert links to the relevant agent detail page.
Agent table
A table of all agents in the current workspace:
| Column | Description |
|---|
| Name | Agent name (links to detail page) |
| Status | Connection + termination status indicators |
| Score | Health score with letter grade |
| Traces/24h | Tool call count in the last 24 hours |
| Risk | Active risk alert level (if any) |
Navigation
The sidebar navigation for the dashboard:
▣ Workspace Overview
🤖 Agents
📊 Tracing
💰 Payments
── Settings ──
⊞ General
👥 Team
🔒 API Keys
🔌 MCP Servers
✉ Billing
Use ⌘K / Ctrl+K to open global search and jump to any agent, workspace, or trace.
