Commit Graph

9 Commits

Author SHA1 Message Date
2fed5d6ce1 Phase 6: Expand market rates and enhance AI investment recommendations
- Rate fetcher now scrapes CD, Money Market, and High Yield Savings rates
  from Bankrate.com with pauses between fetches to avoid rate limiting
- Historical rate data is preserved (no longer deleted on each fetch)
- Database migration adds rate_type column and tenant ai_recommendations table
- Backend returns market rates grouped by type with latest-batch-only queries
- AI prompt now includes all three rate types for comprehensive analysis
- AI recommendations are saved per-tenant for retrieval on page load
- Frontend: "Market CD Rates" replaced with "Today's Market Rates" tabbed view
- Rates section is collapsible (expanded by default) to save screen space
- Saved recommendations load automatically with "Last Updated" timestamp

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 13:39:53 -05:00
a32d4cc179 Add comprehensive platform administration panel
- Database: Add login_history, ai_recommendation_log tables; is_platform_owner
  column on users; subscription fields on organizations (payment_date,
  confirmation_number, renewal_date)
- Backend: New AdminAnalyticsService with platform metrics, tenant detail, and
  health score calculations (0-100 based on activity, budget, transactions,
  members, AI usage)
- Backend: Login/org-switch now records to login_history; AI recommendations
  logged to ai_recommendation_log; platform owner protected from superadmin toggle
- Frontend: 4-tab admin panel (Dashboard, Organizations, Users, Tenant Health)
  with tenant detail drawer, subscription management, health scoring visualization
- Platform owner account (admin@hoaledgeriq.com) auto-redirects to admin panel
- Seed data includes platform owner account and sample login history

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-26 08:51:39 -05:00
0bd30a0eb8 Add 12-month cash flow forecast to AI investment context
The AI was making incorrect liquidity warnings because it only saw
current cash positions and annual budget totals — it had no visibility
into the month-by-month cash flow forecast that includes:
- Special assessment collections (reserve fund income from homeowners)
- Monthly budget income/expense breakdown (not just annual totals)
- Investment maturity schedule (when CDs return cash)
- Capital project expense timing

Now the AI receives:
1. Full assessment schedule (regular + special, with frequency)
2. 12-month forward forecast with projected operating/reserve cash,
   investment balances, and per-month income/expense drivers
3. Updated system prompt instructing the AI to use the forecast for
   liquidity analysis rather than just current balances

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 21:04:57 -05:00
25663fc79e Add AI debug logging and switch from fetch to https module
- Add toggle-able debug logging (AI_DEBUG env var) that logs prompts,
  request metadata, raw responses, parsed output, and full error chains
- Replace Node.js native fetch() with https module for Docker Alpine
  compatibility (fixes "fetch failed" error with large payloads)
- Reduce max_tokens from 16384 to 4096 (qwen3.5 doesn't need thinking
  token budget)
- Strip <think> blocks from model responses
- Add AI_DEBUG to docker-compose.yml and .env.example

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 20:50:39 -05:00
fe4989bbcc Switch AI model from kimi-k2.5 to qwen/qwen3.5-397b-a17b
kimi-k2.5 is a thinking model that times out on complex prompts (>3min).
qwen3.5-397b-a17b responds in ~2s with clean JSON output.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 20:37:48 -05:00
36271585d9 Fix AI timeout and token limits for thinking model (kimi-k2.5)
- Increase max_tokens from 4096 to 16384 to accommodate reasoning tokens
- Increase timeout from 90s to 180s for thinking model latency
- Add logging for response diagnostics (content length, reasoning, finish reason)
- Better error message when model exhausts tokens on reasoning

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 15:50:53 -05:00
18c7989983 Fix: units table uses status='active' not is_active column
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 15:37:47 -05:00
c28d7aeffc Fix TypeScript: export interfaces for controller return type resolution
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 15:32:55 -05:00
f7e9c98bd9 Phase 5: AI investment planning - CD rate fetcher and AI recommendation engine
- Add shared.cd_rates table for cross-tenant market data (CD rates from Bankrate)
- Create standalone Puppeteer scraper script (scripts/fetch-cd-rates.ts) for cron-based rate fetching
- Add investment-planning backend module with 3 endpoints: snapshot, cd-rates, recommendations
- AI service gathers tenant financial data (accounts, investments, budgets, projects, cash flow) and calls OpenAI-compatible API (NVIDIA endpoint) for structured investment recommendations
- Create InvestmentPlanningPage with summary cards, current investments table, market CD rates table, and AI recommendation accordion
- Add Investment Planning to sidebar under Planning menu
- Configure AI_API_URL, AI_API_KEY, AI_MODEL environment variables

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-02-25 15:31:32 -05:00