mirror of
https://github.com/temetro/temetro.git
synced 2026-07-28 04:38:57 +00:00
Khalid Abdi
3a7378e00d
Real LLM chat replacing the mock, backend-centric per the plan: - Multi-provider API-key mode (OpenAI / Anthropic / Gemini via the AI SDK) plus local Ollama (OpenAI-compatible endpoint). Provider is derived from the picked model id; the matching stored key is used. New user_ai_settings table holds per-user config with provider API keys encrypted at rest (AES-256-GCM, src/lib/crypto.ts, keyed by AI_CREDENTIALS_KEY). - POST /api/ai/config (get/put, secrets never returned), POST /api/ai/test (Ollama ping / key presence), POST /api/ai/import (approved migration commit, re-validated server-side, reuses the audited patient service). - POST /api/chat: streamText agent with tools (getPatient, getPatientLabs, searchPatients, previewImport). Real record data streams to the clinician as custom data parts (cards) while the model sees only Veil-redacted results. - Veil (src/services/ai/veil.ts): de-identifies patient identifiers to tokens before external calls, resolves tokens on tool args, and rehydrates the final answer. Bypassed for local Ollama. External mode runs non-streamed so the rehydrated text is correct. Every call is audited (provider + Veil level). - Shared role-scoping helpers extracted to src/lib/role-scope.ts (reused by the patient routes and chat tools so visibility rules match). Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
37 lines
1.7 KiB
Bash
37 lines
1.7 KiB
Bash
# --- Database -------------------------------------------------------------
|
|
# When running via docker compose, "db" is the Postgres service hostname.
|
|
# For running the backend directly on your host, use localhost:5432.
|
|
DATABASE_URL=postgres://temetro:temetro@db:5432/temetro
|
|
|
|
# --- Better Auth ----------------------------------------------------------
|
|
# Generate a strong secret (min 32 chars): openssl rand -base64 32
|
|
BETTER_AUTH_SECRET=replace-me-with-openssl-rand-base64-32
|
|
# Public base URL of THIS backend (used for auth callbacks & cookies).
|
|
BETTER_AUTH_URL=http://localhost:4000
|
|
|
|
# --- AI ------------------------------------------------------------------
|
|
# Key used to encrypt at-rest AI provider API keys (per-user, set in the app's
|
|
# Settings → AI). Generate one: openssl rand -base64 32. Rotating it forces
|
|
# users to re-enter their provider keys.
|
|
AI_CREDENTIALS_KEY=replace-me-with-openssl-rand-base64-32
|
|
|
|
# --- App ------------------------------------------------------------------
|
|
# The frontend origin — used for CORS, Better Auth trustedOrigins, and the
|
|
# links embedded in verification / reset / invitation emails.
|
|
FRONTEND_URL=http://localhost:3000
|
|
PORT=4000
|
|
NODE_ENV=development
|
|
|
|
# Host port Postgres is published on by docker compose. Change it if 5432 is
|
|
# already in use on your machine (the app still talks to Postgres internally).
|
|
POSTGRES_PORT=5432
|
|
|
|
# --- Email (optional) -----------------------------------------------------
|
|
# If SMTP_HOST is unset, emails (verify/reset/invite) are printed to the
|
|
# server console instead of being sent — zero setup for local development.
|
|
# SMTP_HOST=smtp.example.com
|
|
# SMTP_PORT=587
|
|
# SMTP_USER=
|
|
# SMTP_PASS=
|
|
# SMTP_FROM="temetro <no-reply@temetro.local>"
|