Spaces:
Sleeping
Sleeping
|
Download README.md from Sebacu/saniplus-chatbot: direct link, hf CLI and curl.
- Browser
- Download file 2.85 kB
-
https://huggingface.co/spaces/Sebacu/saniplus-chatbot/resolve/main/README.md
- Command line
-
hf download hf://spaces/Sebacu/saniplus-chatbot/README.md
-
curl -L -o README.md https://huggingface.co/spaces/Sebacu/saniplus-chatbot/resolve/main/README.md
2.85 kB
metadata
title: SaniPlus Chatbot
emoji: π
colorFrom: green
colorTo: blue
sdk: docker
app_port: 8100
pinned: false
SaniPlus Apotheken β KI-Assistent (LLM-backed)
A working showcase of the scoped pharmacy chatbot, now powered by a real large language model (Claude) behind a secure backend. It keeps the exact safety design from the decision brief: a deterministic classifier gates every message server-side, clinical and emergency questions never reach the model, and the model answers only from SaniPlus's verified knowledge base.
Architecture
Browser (public/index.html)
β POST /api/chat (conversation history + language)
βΌ
Node + Express backend (server.js) β API key lives here, never in the browser
β 1) Deterministic safety classifier (EMERGENCY / CLINICAL / PII / SAFE)
β β’ EMERGENCY β fixed "call 112 + Notdienst" message (no model call)
β β’ CLINICAL β fixed "see a pharmacist" referral (no model call)
β β’ PII β consent-first callback flow
β β’ SAFE β Claude, grounded on the knowledge base (kb.js)
βΌ
Claude (claude-opus-4-8) β streams the answer back as Server-Sent Events
Two safety layers (defense in depth):
- Server-side classifier β clinical/emergency input is answered by a fixed template and never sent to the model.
- System prompt β even on the SAFE path, the model is instructed to refuse medical advice and answer only from the supplied knowledge base.
Run it
# 1. install
npm install
# 2. set your Anthropic API key
$env:ANTHROPIC_API_KEY = "sk-ant-..."
# 3. start
npm start
# β http://localhost:8100
Without a key the server still runs: the safety gate (emergency/clinical/PII) works fully, and SAFE questions return a friendly "no key configured" message instead of a model answer. Useful for demoing the guardrails offline.
Files
| File | Purpose |
|---|---|
server.js |
Express app, safety classifier, SSE chat endpoint, Claude call |
kb.js |
Verified knowledge base (6 branches + services) and the system prompt |
public/index.html |
Chat UI (streaming, multilingual, quick replies, classifier tags) |
Notes for production
- Opening hours are the site's blanket "MoβSa 9β20" statement β verify per branch (the airport branch differs) before going live.
- Add GDPR/HWG compliance: EU data residency, DPA with the vendor, no training on user data, consent capture for callbacks, minimal logging.
- Consider prompt-caching the system prompt (already marked
cache_control), rate limiting, and a real callback/booking integration for the PII path. - The classifier is intentionally tuned to over-refer: a wrongly deferred safe question costs a click; a wrongly answered clinical one is a liability.