name: rag
version: 1.0.0
description: "Document search, retrieval, and vector storage"
author: "Enclava Team"
category: "ai"

# Module lifecycle
enabled: true
auto_start: true
dependencies: []
optional_dependencies:
  - cache

# Module capabilities
provides:
  - "document_storage"
  - "semantic_search"
  - "vector_embeddings"
  - "document_processing"

consumes:
  - "qdrant_connection"
  - "llm_embeddings"
  - "document_parsing"

# API endpoints
endpoints:
  - path: "/rag/collections"
    method: "GET"
    description: "List document collections"
    
  - path: "/rag/upload"
    method: "POST"
    description: "Upload and process documents"
    
  - path: "/rag/search"
    method: "POST"
    description: "Semantic search in documents"
    
  - path: "/rag/collections/{collection_id}/documents"
    method: "GET"
    description: "List documents in collection"

# UI Configuration
ui_config:
  icon: "search"
  color: "#8B5CF6"
  category: "AI & ML"
  
  forms:
    - name: "collection_config"
      title: "Collection Settings"
      fields: ["name", "description", "embedding_model"]
      
    - name: "search_config"
      title: "Search Configuration"
      fields: ["top_k", "similarity_threshold", "rerank_enabled"]

# Permissions
permissions:
  - name: "rag.create"
    description: "Create document collections"
    
  - name: "rag.upload"
    description: "Upload documents to collections"
    
  - name: "rag.search"
    description: "Search document collections"
    
  - name: "rag.manage"
    description: "Manage all collections (admin)"

# Health checks
health_checks:
  - name: "qdrant_connectivity"
    description: "Check Qdrant vector database connection"
    
  - name: "embeddings_service"
    description: "Check LLM embeddings service"
    
  - name: "document_processing"
    description: "Check document parsing capabilities"