Awesome-omni-skill swarm-dashboard
Live real-time dashboard for monitoring AI agent swarms with smart completion detection, clickable agent detail views, idle time tracking, and activity streaming. Auto-launches a beautiful web UI. V6 features embedded HTML, line-by-line JSON parsing, and capybara-inspired colors.
install
source · Clone the upstream repo
git clone https://github.com/diegosouzapw/awesome-omni-skill
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/diegosouzapw/awesome-omni-skill "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/design/swarm-dashboard" ~/.claude/skills/diegosouzapw-awesome-omni-skill-swarm-dashboard && rm -rf "$T"
manifest:
skills/design/swarm-dashboard/SKILL.mdsource content
Swarm Dashboard Skill
Real-time monitoring for AI agent swarms with a beautiful, auto-refreshing web UI.
Quick Start
from swarm_dashboard import launch_dashboard, update_agent_task_id # Launch dashboard url = launch_dashboard( swarm_name="My Project", swarm_dir="/workspace/my-project", agents={ "agent-1": {"role": "Core Architect", "wave": 1, "mission": "Setup"}, "agent-2": {"role": "Backend Dev", "wave": 2, "mission": "Build API"}, } ) # After launching each agent with Task tool, update task ID update_agent_task_id( swarm_dir="/workspace/my-project", agent_id="agent-1", task_id="abc123" # From Task tool response )
Features
- Real-time Updates - Auto-refreshes every 2 seconds
- Smart Completion Detection - Auto-detects finished agents
- Capybara-Inspired UI - Natural color palette with light/dark themes
- Clickable Details - Live activity feed, tools used, files created
- Zero Dependencies - Pure Python stdlib
CLI Usage
# Start server swarm-dashboard serve --port 8080 --swarm-dir /workspace # Launch in background swarm-dashboard launch --name "My Swarm" --dir /workspace # Check status swarm-dashboard status --dir /workspace # Stop swarm-dashboard stop --dir /workspace
Configuration
Environment variables:
- Directory with agent foldersSWARM_DIR
- Directory with task outputsTASK_DIR
- Server port (default 8080)DASHBOARD_PORT
- Display nameSWARM_NAME
API Endpoints
| Endpoint | Description |
|---|---|
| Dashboard HTML |
| Swarm status JSON |
| Agent details |
| Health check |