Repository navigation
feat(ai): implementar servicio FastAPI con adaptador LLM, prompt v1 y… - #48
Open
Maykol0721 wants to merge 4 commits into
Open
Maykol0721 wants to merge 4 commits into
Maykol0721 wants to merge 4 commits into
Conversation
… candado de datos (S1-18) - Microservicio FastAPI en ai-service con endpoints /health, /info y /explain (ADR-02, RF-IA-02 v1.1). - Esquema formal de afirmaciones tipadas: OBSERVACION, INTERPRETACION y RECOMENDACION con validación de finding_ids y kb_fragment_id. - Adaptador de LLM desacoplado con perfiles openrouter_paid, openrouter_free, gemini_free, ollama_local y mock (ADR-10). - Candado de datos en código que bloquea fail-fast con APP_ENV=prod ante perfiles que comparten datos (RES-06). - Plantilla prompt v1 con delimitación estricta de datos no confiables (RNF-SEG-06). - Dockerfile no-root (usuario 10001) y configuración en docker-compose. - Suite de 34 pruebas en pytest cubriendo candado, esquemas, adaptadores y criterios de aceptación.
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Título: mplementar microservicio FastAPI con adaptador LLM, prompt v1 y candado de datos (S1-18)
Descripción: Implementa el microservicio de IA en Python bajo FastAPI para la generación de diagnósticos estructurados conforme a los requisitos RF-IA-02 (v1.1), ADR-02, ADR-10, RES-06 y RNF-SEG-06:
Endpoints GET /health, GET /info y POST /explain.
Esquema Pydantic formal de afirmaciones tipadas (OBSERVACION, INTERPRETACION, RECOMENDACION) con validación de evidencias (finding_ids) y citas a fragmentos curados (kb_fragment_id).
Adaptador de proveedores LLM con soporte para 5 perfiles: openrouter_paid (ZDR y data collection deny), openrouter_free, gemini_free, ollama_local y mock.
Candado de datos en código que bloquea el inicio (fail-fast) si APP_ENV=prod utiliza proveedores gratuitos/con retención de datos.
Prompt v1 con delimitación estricta de datos no confiables (<<<UNTRUSTED_...>>>) y mitigaciones contra prompt injection.
Dockerfile no-root (UID 10001) y configuración en docker-compose.yml.
Cobertura con 34 pruebas en pytest (34/34 en verde) y suite completa de pruebas unitarias y de arquitectura .NET intactas (284/284 en verde).