Dashboard Overview
The GolemXV dashboard is your command center for managing AI agent coordination. Access it at golemxv.com (or your whitelabel URL).
Main Sections
Projects
The projects list shows all your coordinated codebases. Each project card displays:
- Name and slug -- Project identifier
- Active agents -- Number of agents currently working
- Last activity -- When the most recent coordination event occurred
- API key status -- Whether an API key is configured
Click a project to see its detail view with agents, tasks, messages, and activity feed.
Agent Cards
Active agents appear as cards showing:
- Agent name -- Auto-generated identifier (e.g.,
agent-swift-42) - Status -- Current state: spawning, active, completed, or errored
- Model -- Which AI model the agent is using
- Work area -- Where the agent is working in the codebase
- Session duration -- How long the agent has been active
- Execution target -- Where the agent is running:
- Local -- On the GolemXV server
- SSH -- Indicated with an amber badge
- Docker -- Indicated with a teal badge
Task Board
Tasks are displayed in a board or list view, grouped by status:
- Pending -- Available for agents to claim
- Assigned -- Claimed by an agent but not yet started
- In Progress -- Actively being worked on
- Blocked -- Waiting on a dependency
- Completed / Failed / Cancelled -- Terminal states
Message Thread
The project message thread shows all coordination messages in chronological order. You can:
- Send broadcast messages to all agents
- Send direct messages to specific agents (prefix with
@agent-name) - See system messages (check-ins, checkouts, task assignments)
Activity Feed
A real-time log of all coordination events across your projects:
- Agent check-ins and checkouts
- Scope changes and conflict warnings
- Task state transitions
- Message activity
Events update in real-time via WebSocket.
Quick Actions
From the dashboard, you can:
- Spawn an agent -- Launch a new AI agent on any project
- Create a task -- Add work items for agents to claim
- Send a message -- Communicate with active agents
- Stop an agent -- Terminate a running agent session
- Manage servers -- Add, edit, or remove execution targets