Vadym/OVHserver
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
21 commits 1 branch 0 tags 32 MiB
Commit graph

3 commits

Author SHA1 Message Date
SamoilenkoVadym
a1d606d4fb chore: reset Presenton to clean installation 
- Remove LiteLLM proxy configuration
- Reset to default single-container setup
- Clear all custom LLM and image provider settings
- Remove litellm-config.yaml
- Simple .env with minimal configuration

Service is now in clean state ready for fresh configuration

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-11-06 10:13:20 +00:00
SamoilenkoVadym
0f40ef007f feat: fix Presenton Azure OpenAI integration and add project documentation 
- Add CLAUDE.md with comprehensive project documentation for Claude Code
- Fix Presenton to use LiteLLM proxy instead of direct Azure OpenAI calls
- Add LiteLLM proxy service to Presenton docker-compose
- Create .env.example and litellm-config.yaml.example templates
- Update .gitignore to exclude litellm-config.yaml but allow .env.example files
- Presenton now successfully connects to Azure OpenAI via LiteLLM proxy

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-11-06 09:19:27 +00:00
SamoilenkoVadym
117de92709 feat: add Presenton AI presentation generator with Azure OpenAI integration 
Deployed Presenton (AI-powered presentation generator) to production infrastructure.

Features:
- AI-generated presentations from text prompts and documents
- Azure OpenAI GPT-5 for text generation
- Azure OpenAI DALL-E 3 for image generation
- Export to PPTX and PDF formats
- Custom template support with HTML/Tailwind CSS
- Privacy-focused (local data processing)

Infrastructure:
- Container: ghcr.io/presenton/presenton:latest
- Domain: presentation.ai-impress.com
- SSL: Automatic via Traefik + Cloudflare DNS
- Network: traefik-public
- Resources: 1-2 CPU cores, 1-2GB RAM

Configuration:
- LLM: Azure OpenAI Custom endpoint (GPT-5)
- Images: Azure DALL-E 3
- API keys managed via .env (not committed)
- UI key management enabled (CAN_CHANGE_KEYS=true)

Services started:
- Nginx (port 80) - reverse proxy
- Next.js (port 3000) - frontend
- FastAPI (port 8000) - backend API
- MCP Server (port 8001) - Model Context Protocol

Location: /opt/02-core/presenton/

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <noreply@anthropic.com>
 2025-11-05 23:03:18 +00:00