Self-hosted RAG over the Origam Community forum, built on Microsoft Kernel Memory with Pollinations.AI as the OpenAI-compatible LLM/embedding backend.
cd OrigamAI
dotnet runThen open the interactive API UI:
The root / redirects there. Raw OpenAPI JSON: http://localhost:5244/openapi/v1.json.
If you launch the https profile use https://localhost:7035/scalar/v1 instead.
A small SPA in frontend/ consumes the API. It proxies /api to https://localhost:7035 (the API's HTTPS profile).
cd frontend
npm install # first time only
npm run dev # http://localhost:5173To install Olama for emmbedings
irm https://ollama.com/install.ps1 | iex ollama pull bge-m3 curl http://localhost:11434/api/tags