Files
chat_backend/.env.example
T
westfarn d8f5b8ebf2
Deploy Beta / unit-tests (push) Successful in 11s
Unit Tests / test (push) Successful in 10s
Deploy Beta / docker (push) Successful in 30s
Deploy Beta / deploy-beta (push) Successful in 6m49s
Always-on grounded retrieval + role-scoped Ollama models (#62 Phases 1–3) (#65)
## Summary
- Closes Phases 1–3 of [#62](#62) (Phase 4 eval harness left for a follow-up).
- **Accuracy:** Retrieval is decided every turn (`GroundingDecider`, fails open). `FAST` no longer skips search — it only selects `OLLAMA_MODEL_FAST`. Search failures surface an explicit error instead of hallucinating from parametric memory.
- **Search:** Pluggable `services/search/` with **SearxNG primary** + DDGS failover, ranking/dedupe/rumour filtering, numbered dated source blocks, citations persisted on `Prompt.citations` and emitted as `{"v":1,"type":"citations",...}` after stream end.
- **Models:** Role-scoped `OLLAMA_MODEL_THINKING` / `_FAST` / `_UTILITY` / `OLLAMA_EMBED_MODEL=nomic-embed-text`, configurable `num_ctx`, real model name on `PromptMetric`, `reindex_embeddings` management command + loud embedding-dimension mismatch.

## SearxNG (ops)
See README **SearxNG** section. Short version: run `searxng/searxng` on the GPU host, enable `json` in `settings.yml`, set `SEARXNG_BASE_URL=http://10.0.0.128:8080` in prod/beta secrets, open `:8080` on the LAN firewall like Ollama.

## Test plan
- [x] `SKIP_RAG_INIT=1 python manage.py test chat_backend.tests` — 442 OK (6 skipped)
- [ ] Deploy beta with updated secrets (`OLLAMA_MODEL_*`, `OLLAMA_EMBED_MODEL=nomic-embed-text`, `SEARXNG_BASE_URL`)
- [ ] After embed change: `python manage.py reindex_embeddings`
- [ ] Verify `did Taylor Swift get married` in FAST and THINKING returns grounded answer with citations frame
- [ ] Kill SearxNG and confirm factual turns return search_unavailable (not Joe Alwyn hallucination); non-factual chat still worksReviewed-on: #65
2026-08-02 11:46:02 -07:00

104 lines
4.5 KiB
Bash

# Local development environment file (copy to .env).
# For production server values, use .env.prod.example instead.
DJANGO_ENV=dev
DJANGO_DEBUG=true
DJANGO_SECRET_KEY=change-me-for-local-development
DJANGO_ALLOWED_HOSTS=localhost,127.0.0.1,0.0.0.0
# Optional; when unset, http:// origins are derived for local hosts.
# DJANGO_CSRF_TRUSTED_ORIGINS=http://localhost:8003,http://127.0.0.1:8003
# Database (docker-compose sets DATABASE_URL for the web service)
DATABASE_URL=postgres://chat_backend:chat_backend@db:5432/chat_backend
# Ollama — local loopback when Ollama runs on this machine; LAN IP for GPU host.
OLLAMA_BASE_URL=http://127.0.0.1:11434
# Legacy fallback (used when role-specific vars unset). Prefer the role vars.
# OLLAMA_MODEL=gpt-oss:20b
# OLLAMA_MODEL_THINKING=gpt-oss:20b
# OLLAMA_MODEL_FAST=gemma4:latest
# OLLAMA_MODEL_UTILITY=llama3.2
# OLLAMA_EMBED_MODEL=nomic-embed-text
# OLLAMA_NUM_CTX_THINKING=16384
# OLLAMA_NUM_CTX_FAST=8192
# OLLAMA_NUM_CTX_UTILITY=4096
# Web search (#62) — SearxNG primary, DDGS failover. See README "SearxNG".
ALLOW_INTERNET_ACCESS=true
SEARCH_PROVIDER=searxng
SEARCH_FAILOVER_PROVIDER=ddgs
SEARXNG_BASE_URL=http://127.0.0.1:8088
# SEARXNG_TIMEOUT_SECONDS=8
# Email (SMTP2GO) — optional for local
EMAIL_HOST=mail.smtp2go.com
EMAIL_HOST_USER=
EMAIL_HOST_PASSWORD=
EMAIL_PORT=2525
EMAIL_USE_TLS=true
# Captcha (optional local)
CAPTCHA_SECRET_KEY=
# Self-serve sign-up (default false — set true to allow /user/create/)
ENABLE_ACCOUNT_REGISTRATION=false
# OAuth SSO — Google / Microsoft (#24). Leave blank to hide SSO buttons.
# Redirect URIs (register in each IdP console):
# {OAUTH_CALLBACK_BASE_URL}/api/auth/oauth/google/callback/
# {OAUTH_CALLBACK_BASE_URL}/api/auth/oauth/microsoft/callback/
# Same client id/secret pair is reused for Drive linking (#47) — the extra
# Drive scopes below are requested incrementally via intent=link_drive /
# intent=link_company_drive, no separate app registration needed:
# Google: openid email profile https://www.googleapis.com/auth/drive.readonly
# Microsoft: openid email profile offline_access Files.Read (personal)
# openid email profile offline_access Files.Read.All Sites.Read.All (company)
GOOGLE_OAUTH_CLIENT_ID=
GOOGLE_OAUTH_CLIENT_SECRET=
MICROSOFT_OAUTH_CLIENT_ID=
MICROSOFT_OAUTH_CLIENT_SECRET=
MICROSOFT_OAUTH_TENANT=common
# Optional; defaults to request host. Example local: http://127.0.0.1:8001
OAUTH_CALLBACK_BASE_URL=http://127.0.0.1:8001
# Drive / RAG sync (#47-#53). Requires a subscription plan with allows_rag
# (Founders, Pro, Business, Backer by default — see finance PLAN_SEED).
# Start a link: GET /api/auth/oauth/google/start/?intent=link_drive (authenticated)
# GET /api/auth/oauth/google/start/?intent=link_company_drive (company manager)
# GET /api/auth/oauth/microsoft/start/?intent=link_drive
# GET /api/auth/oauth/microsoft/start/?intent=link_company_drive
# Manage: GET /api/drive/connections/
# DELETE /api/drive/connections/<id>/
# POST /api/drive/connections/<id>/resources/ { "resource_ids": [...] }
# POST /api/drive/connections/<id>/sync/
# Provider push notifications (best-effort; register with each provider's
# subscription/watch API pointing here, using ?connection_id=<id>):
# POST {OAUTH_CALLBACK_BASE_URL}/api/drive/webhooks/google/
# POST {OAUTH_CALLBACK_BASE_URL}/api/drive/webhooks/microsoft/
# Worker sync: `python manage.py sync_drive_connections [--connection-id N]`
# Stripe / finance (optional local — required for checkout + webhooks)
STRIPE_SECRET_KEY=
STRIPE_PUBLISHABLE_KEY=
STRIPE_WEBHOOK_SECRET=
# Optional: pre-created Stripe Price ID for Founders. When empty, Checkout uses
# SubscriptionPlan.price_cents / SUBSCRIPTION_PRICE_* ($10 USD / month Founders).
STRIPE_PRICE_ID=
# SUBSCRIPTION_PRICE_AMOUNT_CENTS=1000
# SUBSCRIPTION_PRICE_CURRENCY=usd
# SUBSCRIPTION_PRICE_INTERVAL=month
# SUBSCRIPTION_PRODUCT_NAME=Founders
# Enforce plan feature + prompt/token quotas on chat turns (default true).
# ENFORCE_SUBSCRIPTION_GATES=true
FRONTEND_BASE_URL=http://localhost:3000
# STRIPE_CHECKOUT_SUCCESS_URL=http://localhost:3000/billing/success?session_id={CHECKOUT_SESSION_ID}
# STRIPE_CHECKOUT_CANCEL_URL=http://localhost:3000/billing/cancel
# Customer Portal return URL (plan change / cancel / payment method).
# STRIPE_PORTAL_RETURN_URL=http://localhost:3000/account/
# Gunicorn / ASGI
GUNICORN_WORKERS=2
GUNICORN_BIND=0.0.0.0:8000
# Host port for docker-compose.prod.yml
WEB_PORT=8003