Skip to content

tuxclaw/engram-dashboard

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 

Repository files navigation

🧠 Engram Dashboard

A native desktop dashboard for Engram — a temporal knowledge graph memory system by Atomlaunch.

This project is a Tauri 2.x desktop app that wraps Engram's FastAPI dashboard with a native window, themed UI, and auto-managed backend. We didn't build Engram itself — just a nice Rust desktop shell with dark themes for exploring the graph.

Tauri Python Neo4j

What This Adds

  • Native desktop window — no browser tab needed
  • Auto-managed backend — Rust spawns the Python server on launch, kills it on close
  • Themes — Dark (default), Dracula, and Nord color schemes
  • Styled dropdowns & UI — WebKitGTK-friendly dark theme overrides

Features (from Engram)

  • Overview — Stats, node type breakdown, recent activity timeline
  • Graph Explorer — Interactive node graph (Sigma.js), click to explore neighbors, filter by type
  • Browse — Search and filter entities, facts, episodes, emotions
  • Multi-agent support — Filter by agent (Andy, Jarvis, etc.)

Architecture

┌─────────────────────────┐
│   Tauri (Rust)          │
│   - Spawns Python       │
│   - Waits for ready     │
│   - WebView → :3460     │
└────────┬────────────────┘
         │
┌────────▼────────────────┐
│   FastAPI (Python)      │
│   - Engram dashboard    │
│   - REST API + static   │
│   - Port 3460           │
└────────┬────────────────┘
         │
┌────────▼────────────────┐
│   Neo4j (Graph DB)      │
│   - Engram data store   │
└─────────────────────────┘

Prerequisites

  • Rust (stable) + Cargo
  • Python 3.10+ with pip
  • Neo4j 5.x running on bolt://localhost:7687
  • Engramgithub.com/Atomlaunch/engram installed and configured
  • System libs (Linux): libwebkit2gtk-4.1-dev, libgtk-3-dev, etc. (Tauri prerequisites)

Setup

1. Install Engram

Follow the setup at github.com/Atomlaunch/engram. Make sure Neo4j is running and Engram's config.json is configured.

2. Install Python dependencies

pip install fastapi uvicorn neo4j

3. Build & Run

git clone https://github.com/tuxclaw/engram-dashboard.git
cd engram-dashboard

# Build the Tauri app
cargo tauri build

# Binary at: src-tauri/target/release/engram-tauri

4. Install (Linux)

cp src-tauri/target/release/engram-tauri ~/.local/bin/engram

API Endpoints (Engram Dashboard)

Endpoint Description
GET /api/status Backend health check
GET /api/stats Node/relationship counts
GET /api/agents List agents in the graph
GET /api/graph Full graph data (nodes + edges)
GET /api/graph/neighbors/{id} Neighbor subgraph
GET /api/search?q=... Full-text search
GET /api/node/{id} Node detail
GET /api/timeline Recent episodes timeline

Tech Stack

  • Desktop: Tauri 2.x (Rust)
  • Backend: FastAPI + uvicorn (Python) — from Engram
  • Database: Neo4j 5.x (Bolt protocol) — from Engram
  • Frontend: Vanilla HTML/JS/CSS, Sigma.js v2 + Graphology
  • Icons: Font Awesome 6
  • Fonts: Inter (Google Fonts)

Credits

  • Engram by Atomlaunch — the memory system and dashboard backend
  • This repo — Tauri desktop wrapper, theme system, and UI polish by tuxclaw

License

MIT

About

Engram — Memory Dashboard (Tauri desktop app)

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors