fix: add missing uuid id generation for new Google OAuth users

fix: correct avatar_url field name in Google OAuth callback; add pool_pre_ping for Neon connection stability

fix: add pool_pre_ping and pool_recycle to prevent stale Neon connection errors

chore: clean up diagnostic print statements from main.py

debug: add print checks for prometheus.yml contents

debug: add GRAFANA_API_KEY diagnostic prints

fix: use request.url.hostname in proto middleware to resolve AttributeError

fix(observability): create prometheus multiproc dir automatically on startup

fix(auth): add fallback profile fetch from Google userinfo API

debug: add startup diagnostics for Google OAuth credentials

fix(auth): implement manual state restoration fallback for OAuth callback

fix(auth): enable same_site none and https_only for cross-domain OAuth cookies

fix(auth): trust reverse proxy headers and force HTTPS scheme in production

fix(agent): handle non-dict LLM response gracefully and delay setting task to FAILED during Celery retries

feat(api): filter cases by authenticated user and protect endpoints

feat(api): add endpoint to list past cases

feat(auth): implement Google OAuth and Email/Password authentication

clean up celery_worker.py by removing rate-limit delays and logs for paid Gemini key

revert LLMService back to official Google GenAI SDK

optimize search result limit to 2 and sleep delay to 12s for OpenRouter free tier stability

add 4-second rate-limiting delay between hypothesis evaluations to respect free tier limits

refactor LLMService to route Gemini requests through OpenRouter

integrate ResearchAgent loop and update_hypothesis database updates in Celery worker

implement ResearchAgent with search formulation and synthesis loops

replace duckduckgo-search with ddgs package and implement search tool

add duckduckgo search tool utility for agent web research

feat: installed the duckduckgo-search via uv

fix: enforce PROMETHEUS_MULTIPROC_DIR at metrics layer top of file to fix multiprocess isolation

feat: configure prometheus multiprocess metrics collection for celery and fastapi

feat: add prometheus metric tracking for token counts, agent latency, and failure rates

feat: mount prometheus metrics app on /metrics endpoint in fastapi gateway

feat: setup prometheus and grafana docker services and initialize core observability metrics

chore: upgrade default gemini model to 2.5-flash and remove hardcoded models

feat: migrate decompose route to enqueue async celery jobs with 202 accepted response

feat: setup celery app initialization and async run_case_manager_task task

feat: complete FastAPI gateway routes and entry point integration

feat: implement CaseManagerAgent and BaseAgent harness execution pipeline

feat: implement google gemini llm helper service with structured json configuration

feat: implement shared context manager for Postgres mapping and context state updates

fix: resolve db sslmode string space and foreign key table naming

feat: created neon db connection using sqlalchemy

feat: define shared context data contracts and configure uv project metadata

feat: added files and folders