Merge pull request #17 from Light-Heart-Labs/dream-server-v1-scrubbed

Add Dream Server — turnkey local AI stack

Author: Lightheartdevs

README.md

@@ -31,12 +31,12 @@ curl -fsSL https://raw.githubusercontent.com/Light-Heart-Labs/Lighthouse-AI/main
 
 | Feature | Dream Server | Ollama + WebUI | LocalAI |
 |---------|:---:|:---:|:---:|
-| One-command install | **Yes** | Manual | Manual |
-| Hardware auto-detect | **Yes** | No | No |
-| Voice agents (STT + TTS + WebRTC) | **Yes** | No | Limited |
-| vLLM inference speed | **Yes** | No | No |
-| Workflow automation | **n8n built in** | No | No |
-| Privacy tools | **PII redaction** | No | No |
+| Full-stack one-command install | **LLM + voice + workflows + RAG + privacy** | LLM + chat only | LLM only |
+| Hardware auto-detect + model selection | **Yes** | No | No |
+| Voice agents (STT + TTS + WebRTC) | **Built in** | No | Limited |
+| Inference engine | **vLLM** (continuous batching) | llama.cpp | llama.cpp |
+| Workflow automation | **n8n (400+ integrations)** | No | No |
+| PII redaction / privacy tools | **Built in** | No | No |
 | Multi-GPU | **Yes** | Partial | Partial |
 
 ---

dream-server/DEMO-SCRIPT.md

@@ -324,7 +324,7 @@ If `nvidia-smi` shows no process, the model is running on CPU (very slow). Fix G
 The Privacy Shield has rate limiting to prevent abuse. Default: 100 requests/minute.
 
 To increase:
-1. Edit `privacy-shield-mvp-bundle/.env`
+1. Edit `.env`
 2. Change `RATE_LIMIT_REQUESTS_PER_MINUTE=100`
 3. Restart: `docker compose restart privacy-shield`
 
@@ -469,7 +469,7 @@ Then restart: `docker compose up -d`
 ### Documentation
 - Main README: `dream-server/README.md`
 - Architecture: `docs/ARCHITECTURE.md`
-- API Reference: `docs/API.md`
+- Security: `SECURITY.md`
 
 ### Community
 - GitHub Issues: https://github.com/Light-Heart-Labs/Lighthouse-AI/issues

dream-server/FAQ.md

@@ -181,13 +181,14 @@ docker compose down              # Stop everything
 
 ## Comparison
 
-| Feature | Dream Server | LocalAI | Ollama + WebUI |
+| Feature | Dream Server | Ollama + WebUI | LocalAI |
 |---------|:---:|:---:|:---:|
-| Voice agents | **Yes** | Limited | No |
-| vLLM speed | **Yes** | No | No |
-| One-click setup | **Yes** | Partial | Yes |
-| Workflow automation | **n8n** | No | No |
-| Hardware auto-detect | **Yes** | No | No |
+| Full-stack one-command install | **LLM + voice + workflows + RAG + privacy** | LLM + chat only | LLM only |
+| Hardware auto-detect + model selection | **Yes** | No | No |
+| Voice agents (STT + TTS + WebRTC) | **Built in** | No | Limited |
+| Inference engine | **vLLM** (continuous batching) | llama.cpp | llama.cpp |
+| Workflow automation | **n8n (400+ integrations)** | No | No |
+| PII redaction / privacy tools | **Built in** | No | No |
 | Multi-GPU | **Yes** | Partial | Partial |
 
 ---