Brainery Containers

Docker containers for the Brainery RAG system.

🇮🇹 Italiano | 🇨🇳 中文

Containers

Container	Port	Purpose
crawl4ai	9100	Web scraping
yt-dlp-server	9101	YouTube transcripts
whisper-server	9102	Audio transcription
anythingllm	9103	RAG database

Port range 9100-9103 is workflow-ordered and uses IANA unassigned space to minimize conflicts. Default ports work out-of-box; optional .env override available.

Quick Start

docker-compose up -d

Verify containers are running:

curl http://localhost:9100/health   # crawl4ai
curl http://localhost:9101/health   # yt-dlp-server
curl http://localhost:9102/health   # whisper-server
curl http://localhost:9103/api/ping # anythingllm

Data Management

Brainery uses a mixed strategy to optimize performance and persistence:

📁 Persistent Data (host disk)

• anythingllm-storage: RAG database with documents and embeddings
• whisper-models: Pre-downloaded Whisper models (~150MB)

Location:

• Windows: C:\ProgramData\Docker\volumes\
• Linux: /var/lib/docker/volumes/

💾 Temporary Caches (RAM)

• crawl4ai: Web scraping cache (512MB tmpfs)
• yt-dlp-server: Temporary video cache (1GB tmpfs)

Advantages:

• ✅ High performance (RAM)
• ✅ No disk I/O for caches
• ✅ RAG database preserved across restarts
• ✅ Whisper models not re-downloaded

RAM requirements: ~1.5GB dedicated for caches + models in execution

---

Port Customization

Default ports (9100-9103) work for most users.

If you have a conflict, create a .env file:

cp .env.example .env
# Edit ports in .env file
docker-compose up -d

Available Images

All images are published on Docker Hub:

• tapiocapioca/crawl4ai:latest
• tapiocapioca/yt-dlp-server:latest
• tapiocapioca/whisper-server:latest
• tapiocapioca/anythingllm:latest

Documentation

• Installation Guide - Step-by-step setup
• Usage Examples - Practical examples

Versioning

Uses Semantic Versioning: v1.0.0, v1.1.0, etc.

Repository Structure

• brainery-containers: This repository (Docker containers)
• brainery: Main Claude Code skill

License

MIT