rckrdmrd
4f5fb5b656
- Add PLAN-DESARROLLO-PENDIENTE.md for agent continuity - Documents current state (77 tables, 9 schemas) - Lists pending tasks by module - Provides execution order recommendations - Includes environment configuration - Update _MAP.md with new plan reference - Update VALIDACION-ALINEACION-ML to 8 phases 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
232 lines
6.5 KiB
Markdown
232 lines
6.5 KiB
Markdown
---
|
|
id: "PLAN-DESARROLLO-PENDIENTE"
|
|
title: "Plan de Desarrollo Pendiente - Trading Platform"
|
|
type: "Plan"
|
|
status: "In Progress"
|
|
priority: "Alta"
|
|
project: "trading-platform"
|
|
version: "1.0.0"
|
|
created_date: "2026-01-07"
|
|
updated_date: "2026-01-07"
|
|
author: "Claude-Orquestador"
|
|
---
|
|
|
|
# Plan de Desarrollo Pendiente - Trading Platform
|
|
|
|
**Última actualización:** 2026-01-07
|
|
**Estado:** En Progreso
|
|
**Para:** Agentes futuros que continúen el desarrollo
|
|
|
|
---
|
|
|
|
## Resumen del Estado Actual
|
|
|
|
### Completado (2026-01-07)
|
|
|
|
1. **Base de Datos**
|
|
- ✅ 9 schemas creados (auth, education, financial, trading, investment, ml, llm, audit, market_data)
|
|
- ✅ 77 tablas, 104 FK
|
|
- ✅ market_data con 6 activos (XAUUSD, EURUSD, BTCUSD, GBPUSD, USDJPY, AUDUSD)
|
|
- ✅ 4M+ registros 5m, 1.3M+ registros 15m (10 años: 2015-2025)
|
|
|
|
2. **ML Engine**
|
|
- ✅ 15 endpoints documentados en ET-ML-004-api.md
|
|
- ✅ 14 modelos en ML_INVENTORY.yml (ML-005 a ML-018)
|
|
- ✅ Backend clients actualizados (ml-engine.client.ts, ml-integration.service.ts)
|
|
- ✅ Trazabilidad completa en TRACEABILITY.yml
|
|
|
|
3. **Documentación**
|
|
- ✅ NEXUS v3.4 frontmatter aplicado
|
|
- ✅ _MAP.md actualizado con todos los schemas
|
|
- ✅ VALIDACION-ALINEACION-ML-2026-01-07.md (8 fases completadas)
|
|
|
|
---
|
|
|
|
## Tareas Pendientes por Módulo
|
|
|
|
### 1. Base de Datos (OQI-009)
|
|
|
|
| Tarea | Prioridad | Descripción |
|
|
|-------|-----------|-------------|
|
|
| Instalar pgvector | Alta | Requerido para llm.embeddings |
|
|
| Crear indices adicionales | Media | Optimizar queries frecuentes |
|
|
| Configurar particionamiento | Media | Para predictions y audit_logs |
|
|
| Implementar RLS | Media | Row Level Security |
|
|
|
|
**Comando para continuar:**
|
|
```bash
|
|
cd apps/database
|
|
./scripts/drop-and-recreate-database.sh
|
|
./scripts/migrate_6_assets.sh
|
|
```
|
|
|
|
### 2. Backend (OQI-002)
|
|
|
|
| Tarea | Prioridad | Estado |
|
|
|-------|-----------|--------|
|
|
| Unificar ML clients | Media | Pendiente - Dos clientes incompatibles |
|
|
| WebSocket signals | Alta | Implementar canal de signals en tiempo real |
|
|
| Auth middleware | Alta | Completar integración con auth schema |
|
|
| API tests | Media | Aumentar cobertura |
|
|
|
|
**Archivos relevantes:**
|
|
- `apps/backend/src/shared/clients/ml-engine.client.ts`
|
|
- `apps/backend/src/modules/ml/services/ml-integration.service.ts`
|
|
|
|
### 3. ML Engine (OQI-006)
|
|
|
|
| Tarea | Prioridad | Estado |
|
|
|-------|-----------|--------|
|
|
| Entrenar modelos BTCUSD | Completado | Actualizado 2026-01-07 |
|
|
| Agregar más activos | Baja | Solo 6 de 17+ disponibles |
|
|
| Optimizar backtesting | Media | Mejorar performance |
|
|
| Implementar monitoring | Alta | Métricas de producción |
|
|
|
|
**Modelos implementados:**
|
|
- ML-005: AttentionScoreModel
|
|
- ML-006: SymbolTimeframeModel
|
|
- ML-007: AssetMetamodel
|
|
- ML-008 a ML-018: Modelos extendidos
|
|
|
|
### 4. Frontend (OQI-003)
|
|
|
|
| Tarea | Prioridad | Descripción |
|
|
|-------|-----------|-------------|
|
|
| Dashboard ML | Alta | Mostrar signals y predicciones |
|
|
| Charts con overlays | Alta | Integrar ML overlays |
|
|
| Configuración de alertas | Media | UI para notificaciones |
|
|
|
|
### 5. LLM Agent (OQI-007)
|
|
|
|
| Tarea | Prioridad | Descripción |
|
|
|-------|-----------|-------------|
|
|
| Instalar pgvector | Alta | Prerrequisito |
|
|
| Configurar embeddings | Alta | RAG implementation |
|
|
| Integrar con ML Engine | Media | Tool: get_signal |
|
|
|
|
---
|
|
|
|
## Orden de Ejecución Recomendado
|
|
|
|
```
|
|
1. [ALTA] Instalar pgvector para LLM
|
|
└── sudo apt install postgresql-16-pgvector
|
|
|
|
2. [ALTA] Ejecutar recreación limpia de BD
|
|
└── ./scripts/drop-and-recreate-database.sh
|
|
└── ./scripts/migrate_6_assets.sh
|
|
|
|
3. [ALTA] Verificar backend compila y tests pasan
|
|
└── cd apps/backend && npm run build && npm test
|
|
|
|
4. [ALTA] Verificar ML Engine funciona
|
|
└── cd apps/ml-engine && python -m pytest
|
|
|
|
5. [MEDIA] Unificar ML clients en backend
|
|
└── Merge ml-engine.client.ts y ml-integration.service.ts
|
|
|
|
6. [MEDIA] Implementar WebSocket signals
|
|
└── apps/backend/src/modules/trading/trading-stream.service.ts
|
|
```
|
|
|
|
---
|
|
|
|
## Archivos Clave para Desarrollo
|
|
|
|
### Base de Datos
|
|
- `apps/database/scripts/create-database.sh` - Script principal
|
|
- `apps/database/schemas/_MAP.md` - Índice de schemas
|
|
- `apps/database/ddl/schemas/market_data/` - DDL datos de mercado
|
|
|
|
### Backend
|
|
- `apps/backend/src/shared/clients/ml-engine.client.ts` - Cliente ML
|
|
- `apps/backend/src/modules/ml/` - Módulo ML
|
|
|
|
### ML Engine
|
|
- `apps/ml-engine/src/api/main.py` - FastAPI endpoints
|
|
- `apps/ml-engine/config/database.yaml` - Configuración BD
|
|
|
|
### Documentación
|
|
- `docs/99-analisis/VALIDACION-ALINEACION-ML-2026-01-07.md` - Validación ML
|
|
- `docs/02-definicion-modulos/OQI-006-ml-signals/` - Especificaciones ML
|
|
|
|
---
|
|
|
|
## Configuración de Ambiente
|
|
|
|
### Variables de Entorno Requeridas
|
|
|
|
```bash
|
|
# PostgreSQL
|
|
DB_HOST=localhost
|
|
DB_PORT=5432
|
|
DB_NAME=trading_platform
|
|
DB_USER=trading_user
|
|
DB_PASSWORD=trading_dev_2025
|
|
|
|
# ML Engine
|
|
ML_ENGINE_URL=http://localhost:3083
|
|
ML_ENGINE_API_KEY=<opcional>
|
|
|
|
# MySQL (solo lectura - datos históricos)
|
|
MYSQL_HOST=72.60.226.4
|
|
MYSQL_USER=root
|
|
MYSQL_PASSWORD=AfcItz2391,.
|
|
MYSQL_DB=db_trading_meta
|
|
```
|
|
|
|
### Puertos
|
|
|
|
| Servicio | Puerto | Descripción |
|
|
|----------|--------|-------------|
|
|
| PostgreSQL | 5432 | Base de datos principal |
|
|
| Backend | 3081 | API NestJS |
|
|
| ML Engine | 3083 | FastAPI ML |
|
|
| Frontend | 3080 | React/Vite |
|
|
|
|
---
|
|
|
|
## Commits Recientes Relevantes
|
|
|
|
| Subproyecto | Commit | Descripción |
|
|
|-------------|--------|-------------|
|
|
| database | `35fb51a` | Add market_data schema + 6 assets |
|
|
| backend | `14da1f1` | Update ML clients with correct API routes |
|
|
| ml-engine | `7d61d54` | Add tests for DirectionalFilters |
|
|
| main | `3f64a12` | Update validation report Phase 8 |
|
|
|
|
---
|
|
|
|
## Notas para Agente Continuador
|
|
|
|
1. **Antes de comenzar**: Verificar estado de BD con:
|
|
```bash
|
|
PGPASSWORD=trading_dev_2025 psql -h localhost -U trading_user -d trading_platform \
|
|
-c "SELECT schemaname, COUNT(*) FROM pg_tables WHERE schemaname NOT IN ('pg_catalog', 'information_schema') GROUP BY schemaname;"
|
|
```
|
|
|
|
2. **Si BD está vacía**: Ejecutar recreación completa:
|
|
```bash
|
|
cd apps/database
|
|
./scripts/drop-and-recreate-database.sh
|
|
./scripts/migrate_6_assets.sh
|
|
```
|
|
|
|
3. **Si hay errores de locale**: El script tiene fallback automático para WSL2
|
|
|
|
4. **Si pgvector falta**: Solo afecta llm.embeddings, resto funciona
|
|
|
|
5. **Documentación**: Siempre actualizar _MAP.md al modificar schemas
|
|
|
|
---
|
|
|
|
## Changelog
|
|
|
|
| Fecha | Versión | Cambios |
|
|
|-------|---------|---------|
|
|
| 2026-01-07 | 1.0.0 | Creación inicial con estado actual |
|
|
|
|
---
|
|
|
|
*Creado: 2026-01-07 | Agente: Claude-Orquestador*
|