- FASE-0: Diagnostic audit of 500+ files, 33 findings cataloged (7P0/8P1/12P2/6P3) - FASE-1: Resolved 7 P0 critical conflicts (ports, paths, dedup OQI-010/ADR-002, orphan schemas) - FASE-2: Resolved 8 P1 issues (traces, README/CLAUDE.md, DEPENDENCY-GRAPH v2.0, DDL drift, stack versions, DoR/DoD) - FASE-3: Resolved 12 P2 issues (archived tasks indexed, RNFs created, OQI-010 US/RF/ET, AGENTS v2.0) - FASE-4: Purged 3 obsolete docs to _archive/, fixed MODELO-NEGOCIO.md broken ref - FASE-5: Cross-layer validation (DDL→OQI 66%, OQI→BE 72%, BE→FE 78%, Inventories 95%) - FASE-6: INFORME-FINAL, SA-INDEX (18 subagents), METADATA COMPLETED 27/33 findings resolved (82%), 6 P3 deferred to backlog. 18 new files created, 40+ modified, 4 archived. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
49 lines
1.2 KiB
Markdown
49 lines
1.2 KiB
Markdown
---
|
|
id: ET-LTI-001
|
|
title: "Arquitectura de Integracion LLM-Trading"
|
|
type: "Specification"
|
|
status: "Backlog"
|
|
rf_parent: "RF-LTI-001"
|
|
epic: OQI-010
|
|
version: "1.0"
|
|
created_date: "2026-02-06"
|
|
---
|
|
|
|
# ET-LTI-001: Arquitectura de Integracion LLM-Trading
|
|
|
|
## Diagrama
|
|
|
|
```
|
|
LLM Agent (OQI-007)
|
|
|
|
|
+-- Tool Registry (OQI-010)
|
|
| +-- Market Tools -> data-service (port 3084)
|
|
| +-- ML Tools -> ml-engine (port 3083)
|
|
| +-- Trading Tools -> backend/trading (port 3081)
|
|
| +-- Portfolio Tools -> backend/portfolio (port 3081)
|
|
|
|
|
+-- Prompt Engine (OQI-010)
|
|
| +-- Template Selector
|
|
| +-- Context Injector
|
|
| +-- Safety Filter
|
|
|
|
|
+-- Guardrails (OQI-010)
|
|
+-- Confirmation Handler
|
|
+-- Risk Validator
|
|
+-- Audit Logger
|
|
```
|
|
|
|
## Flujo de Ejecucion
|
|
|
|
1. Usuario envia mensaje al LLM
|
|
2. LLM selecciona template apropiado basado en intent
|
|
3. LLM invoca tools necesarios
|
|
4. Safety guardrails validan cada tool call
|
|
5. Resultados formateados y retornados al usuario
|
|
6. Si es operacion de trading: confirmacion antes de ejecutar
|
|
|
|
## Tecnologias
|
|
- Tool framework: function calling del LLM (Ollama/Claude)
|
|
- Backend: Express.js endpoints existentes
|
|
- Comunicacion: HTTP entre servicios
|