Knowledge Graph

{
  "raw": "name: kgrag-mcp-server\nimage: mcp/kgrag_mcp_server\ntype: server\nmeta:\n  category: productivity\n  tags:\n    - knowledge-graph\n    - search\n    - productivity\nabout:\n  title: Knowledge Graph\n  description: |-\n    Manage, ingest, and query structured/unstructured data.\n    Designed to integrate with Neo4j, AWS S3, Redis, Qdrant, and LLMs.\n  icon: https://avatars.githubusercontent.com/u/6775950?v=4\n\nsource:\n  project: https://github.com/gzileni/kgrag_mcp_server\n  branch: main\n  commit: 79ae98a75fb5248a6b574b00d3020b3412e53667\n\nconfig:\n  description: Configure the connection to the Knowledge Graph MCP Server\n  env:\n    - name: LLM_MODEL_TYPE\n      example: ollama\n      value: \"{{kgrag-mcp-server.llm_model_type}}\"\n    - name: LLM_MODEL_NAME\n      example: llama3.2\n      value: \"{{kgrag-mcp-server.llm_model_name}}\"\n    - name: API_KEY\n      example: sk-...\n      value: \"{{kgrag-mcp-server.api_key}}\"\n    - name: LLM_URL\n      example: http://localhost:11434\n      value: \"{{kgrag-mcp-server.llm_url}}\"\n    - name: TEMPERATURE\n      example: \"0.5\"\n      value: \"{{kgrag-mcp-server.temperature}}\"\n\n    # Vectors & Collections\n    - name: VECTORDB_SENTENCE_MODEL\n      example: BAAI/bge-small-en-v1.5\n      value: \"{{kgrag-mcp-server.vectordb_sentence_model}}\"\n    - name: COLLECTION_NAME\n      example: kgrag_data\n      value: \"{{kgrag-mcp-server.collection_name}}\"\n\n    # Model embedding\n    - name: MODEL_EMBEDDING\n      example: nomic-embed-text\n      value: \"{{kgrag-mcp-server.model_embedding}}\"\n    - name: LLM_EMBEDDING_URL\n      example: http://localhost:11434\n      value: \"{{kgrag-mcp-server.llm_embedding_url}}\"\n\n    # AWS\n    - name: AWS_ACCESS_KEY_ID\n      example: AK...\n      value: \"{{kgrag-mcp-server.aws_access_key_id}}\"\n    - name: AWS_SECRET_ACCESS_KEY\n      example: cG....\n      value: \"{{kgrag-mcp-server.aws_secret_access_key}}\"\n    - name: AWS_REGION\n      example: eu-central-1\n      value: \"{{kgrag-mcp-server.aws_region}}\"\n    - name: AWS_BUCKET_NAME\n      example: my_bucket\n      value: \"{{kgrag-mcp-server.aws_bucket_name}}\"\n\n  parameters:\n  type: object\n  properties:\n    llm_model_type:\n      type: string\n      default: ollama\n      description: LLM provider (e.g., ollama, openai)\n    llm_model_name:\n      type: string\n      default: llama3.2\n      description: Name of the LLM model to use\n    api_key:\n      type: string\n      description: API key for the LLM provider (not required for local Ollama)\n    llm_url:\n      type: string\n      default: http://localhost:11434\n      description: Base URL of the LLM provider or gateway (e.g., Ollama/OpenAI proxy)\n    temperature:\n      type: number\n      default: 0.5\n      description: Sampling temperature (range 0–2)\n\n    # Vectors & Collections\n    vectordb_sentence_model:\n      type: string\n      default: BAAI/bge-small-en-v1.5\n      description: Sentence embedding model for the vector database\n    collection_name:\n      type: string\n      default: kgrag_data\n      description: Name of the collection/index in the vector database\n\n    # Model embedding\n    model_embedding:\n      type: string\n      default: nomic-embed-text\n      description: Model used for generating embeddings\n    llm_embedding_url:\n      type: string\n      default: http://localhost:11434\n      description: Endpoint for generating embeddings\n\n    # AWS (optional)\n    aws_access_key_id:\n      type: string\n      description: AWS Access Key ID\n    aws_secret_access_key:\n      type: string\n      description: AWS Secret Access Key\n    aws_region:\n      type: string\n      default: eu-central-1\n      description: AWS region\n    aws_bucket_name:\n      type: string\n      description: AWS S3 bucket name\n  required:\n    - llm_model_type\n    - llm_model_name\n",
  "path": "servers/kgrag-mcp-server/server.yaml"
}