Merge pull request #16 from Light-Heart-Labs/dream-server-v1

Adding Dream Server.  One shot setup for local AI democracy. Oh baby...

Author: Lightheartdevs

.gitignore

@@ -41,3 +41,22 @@ secrets/
 token-spy/data/
 token-spy/*.db
 token-spy/*.sqlite
+
+# ============================================
+# Dream Server runtime/build artifacts
+# ============================================
+dream-server/.env
+dream-server/.env.*
+!dream-server/.env.example
+dream-server/data/
+dream-server/models/
+dream-server/config/openclaw/workspace/
+dream-server/dashboard/node_modules/
+dream-server/dashboard/dist/
+dream-server/preflight-*.log
+dream-server/.current-mode
+dream-server/.profiles
+dream-server/.claude/
+dream-server/.pytest_cache/
+dream-server/tests/__pycache__/
+dream-server/internal/

COLLECTIVE.md

@@ -400,6 +400,8 @@ Each component in this repository maps to a specific architectural role:
 
 The toolkit is the infrastructure layer. The [architecture principles](#architecture-principles) are the design layer. Android-Labs is the application layer.
 
+[**Dream Server**](dream-server/) is how all of this ships to users. It packages vLLM, Open WebUI, voice agents, n8n workflows, RAG, and Privacy Shield into a single installer that auto-detects your GPU and gets everything running in 10 minutes. The toolkit components above are the operational backbone that keeps it stable. Dream Server is the product the Collective built — and the fastest way to get local AI running on your own hardware.
+
 You can use the tools without the architecture. But together, they enable something more than the sum of their parts: a system that runs itself.
 
 ---

README.md

@@ -0,0 +1,24 @@
+# Runtime / secrets
+.env
+.env.*
+
+# Install-time data directories
+data/
+models/
+
+# OpenClaw workspace (runtime state)
+config/openclaw/workspace/
+
+# Dashboard build artifacts
+dashboard/node_modules/
+dashboard/dist/
+
+# Preflight logs
+preflight-*.log
+
+# Internal dev
+internal/
+
+# OS junk
+.DS_Store
+Thumbs.db

dream-server/.gitignore

@@ -0,0 +1,63 @@
+# Contributing to Dream Server
+
+Thanks for wanting to help! Here's how to get involved.
+
+## Reporting Issues
+
+Found a bug? Please open an issue with:
+- Your hardware (GPU, RAM, OS)
+- What you expected to happen
+- What actually happened
+- Logs if relevant (`docker compose logs`)
+
+## Pull Requests
+
+1. Fork the repo
+2. Create a feature branch (`git checkout -b feature/cool-thing`)
+3. Make your changes
+4. Test on your hardware
+5. Submit PR with clear description
+
+## What We're Looking For
+
+**High Value:**
+- New workflow templates (n8n JSON exports)
+- Hardware-specific optimizations
+- Better error messages
+- Documentation improvements
+
+**Good First Issues:**
+- Fix typos in docs
+- Add more troubleshooting cases
+- Improve comments in install.sh
+
+**Harder But Appreciated:**
+- Multi-GPU support improvements
+- New model presets
+- Alternative TTS/STT engines
+
+## Testing Your Changes
+
+```bash
+# Fresh install test
+rm -rf ~/dream-server
+./install.sh --dry-run  # Check what would happen
+./install.sh            # Actually install
+
+# Run the status check
+./status.sh
+```
+
+## Code Style
+
+- Bash: Use ShellCheck. We're not religious about style, just be consistent.
+- YAML: 2-space indent, no tabs.
+- Markdown: Keep it readable. No 80-char wrapping.
+
+## Questions?
+
+Open an issue or find us in Discord.
+
+---
+
+*Your contributions help bring local AI to everyone.*

dream-server/CONTRIBUTING.md

@@ -0,0 +1,136 @@
+# Dream Server Client Demo Script
+
+## Overview
+This script guides you through a hands-on demo of Dream Server. The demo covers hardware detection, local LLM chat, voice capabilities, and workflow automation.
+
+**Prerequisites:** Dream Server installed and running (`./scripts/validate.sh` shows green)
+
+---
+
+## 1. Hardware Overview
+
+**Talking Points:**
+- Show the auto-detected hardware tier
+- Explain how Dream Server optimizes for available resources
+
+**Commands:**
+```bash
+# Show GPU info
+nvidia-smi --query-gpu=name,memory.total --format=csv,noheader
+
+# Check what tier Dream Server detected
+cat .env | grep -E "LLM_MODEL|MAX_CONTEXT"
+```
+
+**Expected Output:**
+GPU model and VRAM, plus the model/context settings Dream Server chose.
+
+---
+
+## 2. Local LLM Chat
+
+**Talking Points:**
+- Fully local inference — data never leaves the machine
+- OpenAI-compatible API — drop-in replacement for existing tools
+
+**Commands:**
+```bash
+# Chat completion via API
+curl -s http://localhost:8000/v1/chat/completions \
+  -H "Content-Type: application/json" \
+  -d '{
+    "model": "local",
+    "messages": [{"role": "user", "content": "What is Dream Server in one sentence?"}],
+    "max_tokens": 100
+  }' | jq '.choices[0].message.content'
+```
+
+**Expected Output:**
+A coherent response from the local Qwen model (1-3 seconds).
+
+**Then:** Open http://localhost:3000 to show the chat UI.
+
+---
+
+## 3. Voice Capabilities (if enabled)
+
+**Talking Points:**
+- Speech-to-text with Whisper
+- Text-to-speech with Kokoro
+- Full voice-to-voice conversations via LiveKit
+
+**Commands:**
+```bash
+# Check voice services are running
+docker compose ps whisper piper
+
+# Test STT (if you have an audio file)
+curl -X POST "http://localhost:9000/asr" \
+  -F "audio_file=@test.wav" \
+  -F "output=json"
+```
+
+**Then:** If LiveKit is enabled, open http://localhost:7880 for the voice playground.
+
+---
+
+## 4. Workflow Automation with n8n
+
+**Talking Points:**
+- Visual workflow builder
+- Pre-built workflows for chat, RAG, voice transcription
+- Integrates with any API
+
+**Commands:**
+```bash
+# Check n8n is running
+curl -s http://localhost:5678/ | head -1
+```
+
+**Then:** Open http://localhost:5678 and show the pre-imported workflows.
+
+---
+
+## 5. RAG Document Q&A (if enabled)
+
+**Talking Points:**
+- Upload documents, ask questions
+- Qdrant vector database for semantic search
+- Answers cite their sources
+
+**Commands:**
+```bash
+# Check RAG services
+docker compose ps qdrant embeddings
+```
+
+**Demo flow:** Import the document-qa workflow in n8n, upload a PDF, ask questions.
+
+---
+
+## 6. The Numbers
+
+**Talking Points:**
+- **Cost:** $0/month after hardware (vs $15K+/month for 1M cloud requests)
+- **Latency:** 1.5-2s for 32B model
+- **Capacity:** 30-40 concurrent voice sessions on dual-GPU
+- **Privacy:** Data never leaves your premises
+
+---
+
+## Conclusion
+
+**Summary:**
+- Full-featured AI stack: chat, voice, workflows, RAG
+- Runs on hardware they already own (or can buy once)
+- OpenAI-compatible API — existing tools just work
+- Total data sovereignty
+
+**Next Steps:**
+- Hardware purchase guide: `docs/HARDWARE-GUIDE.md`
+- Pricing tiers: `docs/PRICING-TIERS.md`
+- Contact for install support
+
+---
+
+*Built by The Collective — Android-17, Todd, and friends*