Visual guides and demo walkthroughs for AutoBot's features.
The fastest way to see AutoBot in action is the 3-step Docker setup:
1. git clone https://github.com/mrveiss/AutoBot-AI.git
└── Clone the repository
2. cp .env.example .env && docker compose up -d
└── Configure and start all services (~2 minutes)
3. Open http://localhost
└── AutoBot dashboard is ready
From first command to working dashboard: under 5 minutes on a modern machine.
Open dashboard → Click "New Chat" → Type a message
Example queries:
"What's the status of my fleet?"
"Summarize the deployment runbook"
"Help me write an Ansible playbook for nginx restart"
AutoBot streams responses in real-time. Attach a knowledge base via the KB icon to ground answers in your own documentation.
Sidebar → Knowledge Bases → New Knowledge Base
→ Name it (e.g. "Production Runbooks")
→ Upload documents (PDF, Markdown, text, code files)
→ Documents are indexed in seconds
Then in Chat:
→ Click KB icon → Select "Production Runbooks"
→ Ask questions about your documents
Sidebar → Fleet → Add Node
→ Enter hostname / IP and SSH user
→ Copy AutoBot's SSH public key to the node
→ AutoBot tests connection and enrolls the node
Once enrolled:
→ View real-time CPU / RAM / disk per node
→ Run playbooks: update packages, restart services, check disk
→ Or use Chat: "Restart nginx on all web servers"
Sidebar → Workflows → New Workflow
→ Add steps (chat prompt, fleet op, KB query, delay)
→ Connect steps in sequence
→ Save with a name and optional cron schedule
Example — Weekly Health Check:
Step 1: Fleet → Run "check-disk-usage" on all nodes
Step 2: Chat → "Summarize disk usage, flag nodes above 80%"
Step 3: Chat → "Write a Slack-ready status summary"
Full Mermaid diagrams covering system overview, data flows, and deployment topologies:
Key diagrams:
- System Overview — All components and how they connect
- Chat Data Flow — Sequence diagram from user message to streamed response
- Fleet Operation Flow — How a chat command becomes an Ansible playbook run
- Single-node topology — Standard Docker Compose layout
- Multi-node HA topology — Production high-availability setup
To create a terminal recording (e.g. for a blog post or PR description):
# Install tools
sudo apt install asciinema
pip install agg # or: npm install -g @asciinema/agg
# Record
asciinema rec demo.cast
# Inside the recording:
git clone https://github.com/mrveiss/AutoBot-AI.git
cd AutoBot-AI
cp .env.example .env
docker compose up -d
curl http://localhost/api/v1/health
# Exit recording (Ctrl+D or exit)
# Convert to GIF
agg demo.cast demo.gifSuggested recording script for a clean quickstart demo:
echo "=== AutoBot Quickstart Demo ===" && sleep 1
git clone https://github.com/mrveiss/AutoBot-AI.git && sleep 0.5
cd AutoBot-AI && sleep 0.5
cp .env.example .env && echo "Config ready" && sleep 0.5
docker compose up -d && sleep 2
docker compose ps && sleep 1
curl -s http://localhost/api/v1/health | python3 -m json.tool
echo "✓ AutoBot is running at http://localhost"To take clean screenshots for documentation or social media:
- Start AutoBot:
docker compose up -d - Open
http://localhostin a browser - Recommended views to capture:
- Chat interface with a sample conversation
- Knowledge Base document list
- Fleet dashboard with nodes
- Workflow builder canvas
- Use browser devtools to set viewport to 1440×900 for consistent sizing
To embed a GIF in a markdown file:
To embed a Mermaid diagram (renders on GitHub):
```mermaid
graph TB
...
```Have a demo or screenshot that showcases AutoBot well? Contributions welcome:
- Add your demo file to
docs/demos/(GIF, MP4, or PNG) - Add a description to this file under the appropriate section
- Open a PR targeting
Dev_new_gui