Dobby Dashboard
Live queue control for long-running agent work
Loading queue All projects
Live Jobs
Operations
Project Control
Loading projects...
Recent Activity
Cost Breakdown
Cost Today
--
Cost This Week
--
Cost All Time
--
Tokens Today
--
Stats
Completed Today
--
Total Cost Today
--
Total Cost (All)
--
Total Tokens
--
Avg Duration
--
Tasks Today
--
Current Action
Files Touched
Tool Usage
Logs
Details
Timeline
Select a running task to view logs
Scroll to bottom
📖 Dobby Usage Guide
What is Dobby?

Dobby is a meta-agent orchestrator that dispatches coding tasks to AI agents (Claude Code, Codex) across multiple machines. Submit work from Slack, CLI, or Claude Code — agents execute on your Mac, cloud servers, or GPU clusters while you track progress here.

💬 Submit via Slack
Submit a task (mention the project name):
@Dobby fix the login bug in spare @Dobby add docs to dobby with high priority @Dobby run eval on spare-tillicum using codex
Batch tasks (bullet list):
@Dobby
- fix tests in spare
- add docs to dobby
- run eval on adpo
Check status:
status — bullet-point overview (last 4h)
status 12h — custom time range
summary — detailed AI summary (last 12h)
summary spare 24h — per-project with time range
running — currently executing tasks
projects — list all projects with their spoke servers
Free-form chat (no project = Claude conversation):
@Dobby what's the best way to structure a training loop?
🔧 Submit via MCP (Claude Code)
In any Claude Code session with Dobby MCP configured:
dobby_add_task(project="spare", description="fix the training loop", target="mac-local") dobby_status() dobby_projects()
MCP config (~/.claude/.mcp.json):
{"mcpServers": {"dobby": {"command": "uv", "args": ["run", "--project", "/path/to/dobby", "dobby-mcp"], "env": {"DOBBY_URL": "https://your-hub-url"}}}}
💻 Submit via CLI
dobby add -p spare "fix the training loop" dobby status dobby logs t-20260322-abcd
🌐 Submit via API
curl -X POST https://your-hub/tasks -H 'Content-Type: application/json' \
  -d '{"project":"spare", "description":"fix the bug", "target":"mac-local"}'
🖥 Connect a Spoke Agent
On any machine with Claude Code installed:
pip install dobby && dobby agent --hub https://your-hub --id my-machine

The spoke registers with the hub, polls for tasks targeted at it, executes them locally, and streams logs back to this dashboard.

📊 Dashboard Features

Agent Network — see connected spokes with status dots
Heatmap — GitHub-style activity chart per project
Flow View — VibeLens-style agent execution timeline (click any task → Logs → Flow)
Live Jobs — currently running tasks with real-time updates
Usage — Claude rate limit utilization bars