PhotoMind

The most versatile AI-powered photo assistant. Designed for NAS/Docker deployment.

• AI Search — Find photos by natural language or image similarity, powered by embedding vectors
• AI Management — Chat with an AI agent to organize, move, and manage your photos
• AI Chat — Conversational interface for photo operations with tool confirmation workflow
• Universal Compatibility — Integrate with any photo album app via CLI or HTTP API tools

Rust backend + React frontend. Supports Anthropic, Google, OpenAI, and any OpenAI-compatible provider.

Prerequisites

• Rust (1.94+)
• Node.js (22+)
• Build tools: sudo apt install build-essential pkg-config libssl-dev

Quick Start

cd ~/ETProject/PhotoMind

# Install frontend dependencies (first time only)
cd web && npm install && cd ..

# Build and run
make run

Open http://localhost:8080

Run Methods

1. Make (Recommended)

# Development mode — backend debug + frontend HMR hot reload
make dev        # then open http://localhost:5173

# Production mode — build everything and start
make run        # then open http://localhost:8080

# Build only (no start)
make build

# Clean all build artifacts
make clean

2. Manual — Backend Only

# Debug build + run
cargo run

# Or release build + run
cargo build --release
./target/release/photomind

Backend listens on http://localhost:8080. Serves the frontend from web/dist/ (must be built first).

3. Manual — Frontend Only

cd web

# Development server with hot reload (proxies API to :8080)
npm run dev     # http://localhost:5173

# Production build (outputs to web/dist/)
npm run build

4. Docker

# Edit docker-compose.yml to set your photos directory, then:
docker compose up --build

Environment Variables

Variable	Default	Description
PHOTOMIND_DATA_DIR	data	SQLite database + thumbnail cache
PHOTOMIND_ADDR	0.0.0.0:8080	Server listen address
RUST_LOG	info	Log level (debug, info, warn, error)

Setup After First Launch

1. Open Settings page
2. Add Scan Directories — paths to your photo folders
3. Configure Embedding Model — Google API URL + Key, select gemini-embedding
4. Configure Agent Model — choose provider (Anthropic/Google/OpenAI), enter URL + Key, select model
5. Click Save Settings → Scan & Embed
6. Go to Search to find photos, or Chat to talk with the agent

Project Structure

PhotoMind/
├── crates/
│   ├── server/        Axum HTTP server + API routes
│   ├── core/          Scanner, embedding, search, agent, file watcher
│   ├── storage/       SQLite database layer (photos, embeddings, configs, tools, chat)
│   └── tools/         Tool error types
├── web/               React frontend (Vite + TypeScript + Tailwind)
├── Dockerfile         Multi-stage build
├── docker-compose.yml
└── Makefile