Ontology MCP

Instructions

Implementation Reference

• src/services/openai-service.ts:320-369 (handler)

  Core handler implementation: Calls OpenAI Embeddings API (/v1/embeddings) with text input, model, and optional dimensions, returns JSON of response data.

  /**
   * Embeddings API를 사용하여 텍스트 임베딩을 생성합니다
   */
  async generateEmbeddings(args: {
    text: string | string[];
    model?: string;
    dimensions?: number;
  }): Promise<string> {
    try {
      if (!OPENAI_API_KEY) {
        throw new McpError(
          ErrorCode.InternalError,
          'OPENAI_API_KEY가 설정되지 않았습니다.'
        );
      }
  
      const response = await axios.post(
        `${OPENAI_API_BASE}/embeddings`,
        {
          model: args.model || 'text-embedding-3-small',
          input: args.text,
          dimensions: args.dimensions
        },
        {
          headers: {
            'Content-Type': 'application/json',
            'Authorization': `Bearer ${OPENAI_API_KEY}`
          }
        }
      );
  
      return JSON.stringify(response.data, null, 2);
    } catch (error) {
      if (axios.isAxiosError(error)) {
        const statusCode = error.response?.status;
        const responseData = error.response?.data;
        
        throw new McpError(
          ErrorCode.InternalError,
          `OpenAI API 오류 (${statusCode}): ${
            typeof responseData === 'object' 
              ? JSON.stringify(responseData, null, 2) 
              : responseData || error.message
          }`
        );
      }
      
      throw new McpError(ErrorCode.InternalError, `임베딩 생성 요청 실패: ${formatError(error)}`);
    }
  }

• src/tools/index.ts:762-800 (registration)

  Tool registration: Defines name, description, input schema, and thin handler that delegates to openaiService.generateEmbeddings.

    name: 'mcp_openai_embedding',
    description: 'OpenAI Embeddings API를 사용하여 텍스트 임베딩을 생성합니다',
    inputSchema: {
      type: 'object',
      properties: {
        text: {
          type: ['string', 'array'],
          description: '임베딩을 생성할 텍스트 또는 텍스트 배열'
        },
        model: {
          type: 'string',
          description: '사용할 모델 (예: text-embedding-3-small, text-embedding-3-large)'
        },
        dimensions: {
          type: 'number',
          description: '임베딩 차원 수 (API가 지원하는 경우)'
        }
      },
      required: ['text']
    },
    async handler(args: any): Promise<ToolResponse> {
      try {
        const result = await openaiService.generateEmbeddings(args);
        return {
          content: [{
            type: 'text',
            text: result
          }]
        };
      } catch (error) {
        return {
          content: [{
            type: 'text',
            text: `OpenAI 임베딩 오류: ${error instanceof Error ? error.message : String(error)}`
          }]
        };
      }
    }
  },

• src/tools/index.ts:764-781 (schema)

  Input schema validation for the tool: requires 'text', optional 'model' and 'dimensions'.

  inputSchema: {
    type: 'object',
    properties: {
      text: {
        type: ['string', 'array'],
        description: '임베딩을 생성할 텍스트 또는 텍스트 배열'
      },
      model: {
        type: 'string',
        description: '사용할 모델 (예: text-embedding-3-small, text-embedding-3-large)'
      },
      dimensions: {
        type: 'number',
        description: '임베딩 차원 수 (API가 지원하는 경우)'
      }
    },
    required: ['text']
  },

• src/index.ts:43-43 (registration)

  Tool enabled (true) in the main server configuration flags.

  mcp_openai_embedding: true,