Commit Graph

2 Commits

Author SHA1 Message Date
home
7a40ff629e feat: LLM routing by tier (free→Ollama, pro→Timeweb)
Some checks failed
Build and Deploy GooSeek / build-and-deploy (push) Failing after 8m25s
- Add tier-based provider routing in llm-svc
  - free tier → Ollama (local qwen3.5:9b)
  - pro/business → Timeweb Cloud AI
- Add /api/v1/embed endpoint for embeddings via Ollama
- Update Ollama client: qwen3.5:9b default, remove auth
- Add GenerateEmbedding() function for qwen3-embedding:0.6b
- Add Ollama K8s deployment with GPU support (RTX 4060 Ti)
- Add monitoring stack (Prometheus, Grafana, Alertmanager)
- Add Grafana dashboards for LLM and security metrics
- Update deploy.sh with monitoring and Ollama deployment

Made-with: Cursor
2026-03-03 02:25:22 +03:00
home
a0e3748dde feat: auth service + security audit fixes + cleanup legacy services
Major changes:
- Add auth-svc: JWT auth, register/login/refresh, password reset
- Add auth UI: modals, pages (/login, /register, /forgot-password)
- Add usage tracking (usage_metrics table, daily limits)
- Add tiered rate limiting (free/pro/business)
- Add LLM usage limits per tier

Security fixes:
- All repos now require userID for Update/Delete operations
- JWT middleware in chat-svc, llm-svc, agent-svc, discover-svc
- ErrNotFound/ErrForbidden errors for proper access control

Cleanup:
- Remove legacy TypeScript services/ directory
- Remove computer-svc (to be reimplemented)
- Remove old deploy/docker configs

New files:
- backend/cmd/auth-svc/main.go
- backend/internal/auth/{types,repository}.go
- backend/internal/usage/{types,repository}.go
- backend/pkg/middleware/{llm_limits,ratelimit_tiered}.go
- backend/webui/src/components/auth/*
- backend/webui/src/app/(auth)/*

Made-with: Cursor
2026-02-28 01:33:49 +03:00