Interface Overview
Aegis AI is a desktop application with a sidebar-driven navigation and a settings flyout menu. Everything you need is accessible within two clicks.
Sidebar
The left sidebar contains navigation icons for all main views. It's always visible and serves as the primary way to move between areas of the application.
Core Views (Always Visible)
| Icon | View | Description |
|---|---|---|
| Monitor | Monitor | Live camera grid with configurable layouts, drag-and-drop reordering, and pagination |
| Timeline | Timeline | Browse and filter recorded clips with infinite scroll, camera filters, and favorites |
| Aegis Home | Aegis Home | Chat with the AI agent — full conversational interface with camera awareness |
Skill-Unlocked Views
These views appear only when the matching skill is installed from the marketplace:
| Icon | View | Unlocked By | Description |
|---|---|---|---|
| OpenClaw | OpenClaw | Camera Claw skill | Robotic arm control — sandbox, record, and monitor OpenClaw activity |
| Detection | Detection Studio | YOLO Detection skill | Real-time object detection with bounding box overlays on camera feeds. Supports 80+ COCO classes with adjustable confidence thresholds and model sizes. |
| Depth | 3D Depth Vision | Depth Estimation skill | Depth map visualization using Depth Anything v2 — transforms camera feeds with colorized depth maps for privacy-preserving monitoring |
| Segment | Segmentation Studio | SAM2 Segmentation skill | Interactive click-to-segment using Segment Anything 2 — pixel-perfect masks with point/box prompts and video tracking |
| Annotation | Annotation Studio | Annotation Data skill | Image annotation for custom model training — COCO labels, sequences, and Kaggle export |
Bottom Controls
At the bottom of the sidebar you'll find:
- Skills (puzzle piece icon) — opens the Skills panel for installing and managing AI capabilities
- Settings (gear icon) — opens the settings flyout for all configuration
Settings Flyout
Click the ⚙️ gear icon at the bottom of the sidebar. A flyout menu appears with two sections:
Agent
Expandable groups with sub-items that control the agent's identity, capabilities, and camera sources:
Persona — configures the agent's identity and senses
- Soul — personality name, description, and behavioral instructions. The personality text shapes the agent's tone across all interactions — chat, alerts, voice responses, and messaging channel notifications. Learn more →
- LLM — language model selection and configuration. Choose between a built-in local model, OpenAI API, or any OpenAI-compatible endpoint for the agent's reasoning engine. Learn more →
- VLM — vision model selection. This is the model that watches your cameras — it analyzes every frame and produces the natural-language descriptions that power the entire system. Learn more →
- Channels — messaging platform integrations. Connect Telegram, Discord, or Slack to receive alerts and hold two-way conversations with the agent from your phone. Learn more →
- Voice — text-to-speech model and audio output. Choose from browser voices, downloadable AI models (Kokoro, XTTS, Qwen3-TTS), or the Orpheus neural voice. Learn more →
- Memory — view, add, and manage what the agent remembers. Memories are categorized files containing facts about your family, routines, preferences, and environment. Learn more →
Abilities — additional agent capabilities
- Craft — video and image generation settings. Connect Google Vertex AI, Gemini, or OpenAI to generate videos and images on demand via chat. Learn more →
- Skills — the toolbox of actions the agent can invoke during conversations — web search, weather, camera control, memory writing, and any tools added by installed skills. Learn more →
Cameras — camera source management
- Security Cameras — add and configure Blink, Ring, or LAN cameras. Includes the ONVIF scanner, network device scanner, and manual RTSP entry. Learn more →
- Webcam (Builtin & USB) — add webcams, USB cameras, or desktop capture sources with live preview. Learn more →
System
Top-level system settings that control the engine, storage, and privacy:
- AI Engine — manage the local inference engine. Start/stop the engine, view loaded models, and monitor inference performance. Learn more →
- Storage — disk usage metrics, storage modes (Free Space, Max Usage, Fixed Size), retention policies, and custom media paths. Learn more →
- Hardware — comprehensive system diagnostics showing CPU, GPU, RAM, ML framework compatibility, supported precision formats, and personalized recommendations. Learn more →
- Privacy — crash reporting toggle and blind mode controls. Learn more →
Settings Modal
Clicking Storage or Privacy from the flyout opens a full-screen settings modal with a left navigation panel and three tabs:
| Tab | Content |
|---|---|
| Privacy | Crash reporting toggle (anonymous, no footage included), link to the privacy policy, and blind mode controls |
| Storage | Storage overview (total used, available, clip count, model storage), storage mode selector, retention period, custom media path, and usage trend chart |
| Updates | Update mode selector (automatic, manual, or disabled), check for updates button, download progress, and current version number |
Drawers and Modals
Throughout the application, detailed settings open in slide-out drawers or full-screen modals:
- Agent Settings Drawer — opens when you click any sub-item under "Agent" in the settings flyout. It provides a detailed configuration panel for the selected feature (e.g., Soul, LLM, VLM, Channels). You can navigate between tabs within the drawer without closing it.
- Camera Settings Modals — individual camera configuration opens as a modal with stream settings, credentials, and preview controls.
- Skills Panel — the full skills management interface, including the installed skills list, marketplace, live console, source code browser, and approval rules.
- Clip Playback Modal — opens from the Timeline, providing a full video player with frame-by-frame AI analysis.
- Stream Verification Modal — tests multiple RTSP URLs from a discovered camera and shows the first working stream with a live preview.
Keyboard Shortcuts
| Shortcut | Action |
|---|---|
| Space | Play / Pause video in clip playback |
| Escape | Exit fullscreen mode |
| Enter | Send a message in chat |
| Scroll | Navigate through timeline clips (infinite scroll) |
| Drag | Reorder camera tiles in the Monitor grid |