everything.ts

import { Server } from "@modelcontextprotocol/sdk/server/index.js";
import {
  CallToolRequestSchema,
  CompleteRequestSchema,
  CreateMessageRequest,
  CreateMessageResultSchema,
  GetPromptRequestSchema,
  ListPromptsRequestSchema,
  ListResourcesRequestSchema,
  ListResourceTemplatesRequestSchema,
  ListToolsRequestSchema,
  LoggingLevel,
  ReadResourceRequestSchema,
  Resource,
  SetLevelRequestSchema,
  SubscribeRequestSchema,
  Tool,
  ToolSchema,
  UnsubscribeRequestSchema,
} from "@modelcontextprotocol/sdk/types.js";
import { z } from "zod";
import { zodToJsonSchema } from "zod-to-json-schema";

const ToolInputSchema = ToolSchema.shape.inputSchema;
type ToolInput = z.infer<typeof ToolInputSchema>;

/* Input schemas for tools implemented in this server */
const EchoSchema = z.object({
  message: z.string().describe("Message to echo"),
});

const AddSchema = z.object({
  a: z.number().describe("First number"),
  b: z.number().describe("Second number"),
});

const LongRunningOperationSchema = z.object({
  duration: z
    .number()
    .default(10)
    .describe("Duration of the operation in seconds"),
  steps: z.number().default(5).describe("Number of steps in the operation"),
});

const PrintEnvSchema = z.object({});

const SampleLLMSchema = z.object({
  prompt: z.string().describe("The prompt to send to the LLM"),
  maxTokens: z
    .number()
    .default(100)
    .describe("Maximum number of tokens to generate"),
});

// Example completion values
const EXAMPLE_COMPLETIONS = {
  style: ["casual", "formal", "technical", "friendly"],
  temperature: ["0", "0.5", "0.7", "1.0"],
  resourceId: ["1", "2", "3", "4", "5"],
};

const GetTinyImageSchema = z.object({});

const AnnotatedMessageSchema = z.object({
  messageType: z.enum(["error", "success", "debug"])
    .describe("Type of message to demonstrate different annotation patterns"),
  includeImage: z.boolean().default(false)
    .describe("Whether to include an example image")
});

enum ToolName {
  ECHO = "echo",
  ADD = "add",
  LONG_RUNNING_OPERATION = "longRunningOperation",
  PRINT_ENV = "printEnv",
  SAMPLE_LLM = "sampleLLM",
  GET_TINY_IMAGE = "getTinyImage",
  ANNOTATED_MESSAGE = "annotatedMessage",
}

enum PromptName {
  SIMPLE = "simple_prompt",
  COMPLEX = "complex_prompt",
}

export const createServer = () => {
  const server = new Server(
    {
      name: "example-servers/everything",
      version: "1.0.0",
    },
    {
      capabilities: {
        prompts: {},
        resources: { subscribe: true },
        tools: {},
        logging: {},
      },
    },
  );

  let subscriptions: Set<string> = new Set();
  let subsUpdateInterval: NodeJS.Timeout | undefined;
  // Set up update interval for subscribed resources

  subsUpdateInterval = setInterval(() => {
    for (const uri of subscriptions) {
      server.notification({
        method: "notifications/resources/updated",
        params: { uri },
      });
    }
  }, 5000);

  let logLevel: LoggingLevel = "debug";
  let logsUpdateInterval: NodeJS.Timeout | undefined;
  const messages = [
    {level: "debug", data: "Debug-level message"},
    {level: "info", data: "Info-level message"},
    {level: "notice", data: "Notice-level message"},
    {level: "warning", data: "Warning-level message"},
    {level: "error", data: "Error-level message"},
    {level: "critical", data: "Critical-level message"},
    {level: "alert", data: "Alert level-message"},
    {level: "emergency", data: "Emergency-level message"}
  ]

  const isMessageIgnored = (level:LoggingLevel):boolean => {
    const currentLevel = messages.findIndex((msg) => logLevel === msg.level);
    const messageLevel =  messages.findIndex((msg) => level === msg.level);
    return messageLevel < currentLevel;
  }

  // Set up update interval for random log messages
  logsUpdateInterval = setInterval(() => {
    let message = {
      method: "notifications/message",
      params: messages[Math.floor(Math.random() * messages.length)],
    }
    if (!isMessageIgnored(message.params.level as LoggingLevel)) server.notification(message);
  }, 15000);

  // Helper method to request sampling from client
  const requestSampling = async (
    context: string,
    uri: string,
    maxTokens: number = 100,
  ) => {
    const request: CreateMessageRequest = {
      method: "sampling/createMessage",
      params: {
        messages: [
          {
            role: "user",
            content: {
              type: "text",
              text: `Resource ${uri} context: ${context}`,
            },
          },
        ],
        systemPrompt: "You are a helpful test server.",
        maxTokens,
        temperature: 0.7,
        includeContext: "thisServer",
      },
    };

    return await server.request(request, CreateMessageResultSchema);
  };

  const ALL_RESOURCES: Resource[] = Array.from({ length: 100 }, (_, i) => {
    const uri = `test://static/resource/${i + 1}`;
    if (i % 2 === 0) {
      return {
        uri,
        name: `Resource ${i + 1}`,
        mimeType: "text/plain",
        text: `Resource ${i + 1}: This is a plaintext resource`,
      };
    } else {
      const buffer = Buffer.from(`Resource ${i + 1}: This is a base64 blob`);
      return {
        uri,
        name: `Resource ${i + 1}`,
        mimeType: "application/octet-stream",
        blob: buffer.toString("base64"),
      };
    }
  });

  const PAGE_SIZE = 10;

  server.setRequestHandler(ListResourcesRequestSchema, async (request) => {
    const cursor = request.params?.cursor;
    let startIndex = 0;

    if (cursor) {
      const decodedCursor = parseInt(atob(cursor), 10);
      if (!isNaN(decodedCursor)) {
        startIndex = decodedCursor;
      }
    }

    const endIndex = Math.min(startIndex + PAGE_SIZE, ALL_RESOURCES.length);
    const resources = ALL_RESOURCES.slice(startIndex, endIndex);

    let nextCursor: string | undefined;
    if (endIndex < ALL_RESOURCES.length) {
      nextCursor = btoa(endIndex.toString());
    }

    return {
      resources,
      nextCursor,
    };
  });

  server.setRequestHandler(ListResourceTemplatesRequestSchema, async () => {
    return {
      resourceTemplates: [
        {
          uriTemplate: "test://static/resource/{id}",
          name: "Static Resource",
          description: "A static resource with a numeric ID",
        },
      ],
    };
  });

  server.setRequestHandler(ReadResourceRequestSchema, async (request) => {
    const uri = request.params.uri;

    if (uri.startsWith("test://static/resource/")) {
      const index = parseInt(uri.split("/").pop() ?? "", 10) - 1;
      if (index >= 0 && index < ALL_RESOURCES.length) {
        const resource = ALL_RESOURCES[index];
        return {
          contents: [resource],
        };
      }
    }

    throw new Error(`Unknown resource: ${uri}`);
  });

  server.setRequestHandler(SubscribeRequestSchema, async (request) => {
    const { uri } = request.params;
    subscriptions.add(uri);

    // Request sampling from client when someone subscribes
    await requestSampling("A new subscription was started", uri);
    return {};
  });

  server.setRequestHandler(UnsubscribeRequestSchema, async (request) => {
    subscriptions.delete(request.params.uri);
    return {};
  });

  server.setRequestHandler(ListPromptsRequestSchema, async () => {
    return {
      prompts: [
        {
          name: PromptName.SIMPLE,
          description: "A prompt without arguments",
        },
        {
          name: PromptName.COMPLEX,
          description: "A prompt with arguments",
          arguments: [
            {
              name: "temperature",
              description: "Temperature setting",
              required: true,
            },
            {
              name: "style",
              description: "Output style",
              required: false,
            },
          ],
        },
      ],
    };
  });

  server.setRequestHandler(GetPromptRequestSchema, async (request) => {
    const { name, arguments: args } = request.params;

    if (name === PromptName.SIMPLE) {
      return {
        messages: [
          {
            role: "user",
            content: {
              type: "text",
              text: "This is a simple prompt without arguments.",
            },
          },
        ],
      };
    }

    if (name === PromptName.COMPLEX) {
      return {
        messages: [
          {
            role: "user",
            content: {
              type: "text",
              text: `This is a complex prompt with arguments: temperature=${args?.temperature}, style=${args?.style}`,
            },
          },
          {
            role: "assistant",
            content: {
              type: "text",
              text: "I understand. You've provided a complex prompt with temperature and style arguments. How would you like me to proceed?",
            },
          },
          {
            role: "user",
            content: {
              type: "image",
              data: MCP_TINY_IMAGE,
              mimeType: "image/png",
            },
          },
        ],
      };
    }

    throw new Error(`Unknown prompt: ${name}`);
  });

  server.setRequestHandler(ListToolsRequestSchema, async () => {
    const tools: Tool[] = [
      {
        name: ToolName.ECHO,
        description: "Echoes back the input",
        inputSchema: zodToJsonSchema(EchoSchema) as ToolInput,
      },
      {
        name: ToolName.ADD,
        description: "Adds two numbers",
        inputSchema: zodToJsonSchema(AddSchema) as ToolInput,
      },
      {
        name: ToolName.PRINT_ENV,
        description: "Prints all environment variables, helpful for debugging MCP server configuration",
        inputSchema: zodToJsonSchema(PrintEnvSchema) as ToolInput,
      },
      {
        name: ToolName.LONG_RUNNING_OPERATION,
        description:
          "Demonstrates a long running operation with progress updates",
        inputSchema: zodToJsonSchema(LongRunningOperationSchema) as ToolInput,
      },
      {
        name: ToolName.SAMPLE_LLM,
        description: "Samples from an LLM using MCP's sampling feature",
        inputSchema: zodToJsonSchema(SampleLLMSchema) as ToolInput,
      },
      {
        name: ToolName.GET_TINY_IMAGE,
        description: "Returns the MCP_TINY_IMAGE",
        inputSchema: zodToJsonSchema(GetTinyImageSchema) as ToolInput,
      },
      {
        name: ToolName.ANNOTATED_MESSAGE,
        description: "Demonstrates how annotations can be used to provide metadata about content",
        inputSchema: zodToJsonSchema(AnnotatedMessageSchema) as ToolInput,
      },
    ];

    return { tools };
  });

  server.setRequestHandler(CallToolRequestSchema, async (request) => {
    const { name, arguments: args } = request.params;

    if (name === ToolName.ECHO) {
      const validatedArgs = EchoSchema.parse(args);
      return {
        content: [{ type: "text", text: `Echo: ${validatedArgs.message}` }],
      };
    }

    if (name === ToolName.ADD) {
      const validatedArgs = AddSchema.parse(args);
      const sum = validatedArgs.a + validatedArgs.b;
      return {
        content: [
          {
            type: "text",
            text: `The sum of ${validatedArgs.a} and ${validatedArgs.b} is ${sum}.`,
          },
        ],
      };
    }

    if (name === ToolName.LONG_RUNNING_OPERATION) {
      const validatedArgs = LongRunningOperationSchema.parse(args);
      const { duration, steps } = validatedArgs;
      const stepDuration = duration / steps;
      const progressToken = request.params._meta?.progressToken;

      for (let i = 1; i < steps + 1; i++) {
        await new Promise((resolve) =>
          setTimeout(resolve, stepDuration * 1000),
        );

        if (progressToken !== undefined) {
          await server.notification({
            method: "notifications/progress",
            params: {
              progress: i,
              total: steps,
              progressToken,
            },
          });
        }
      }

      return {
        content: [
          {
            type: "text",
            text: `Long running operation completed. Duration: ${duration} seconds, Steps: ${steps}.`,
          },
        ],
      };
    }

    if (name === ToolName.PRINT_ENV) {
      return {
        content: [
          {
            type: "text",
            text: JSON.stringify(process.env, null, 2),
          },
        ],
      };
    }

    if (name === ToolName.SAMPLE_LLM) {
      const validatedArgs = SampleLLMSchema.parse(args);
      const { prompt, maxTokens } = validatedArgs;

      const result = await requestSampling(
        prompt,
        ToolName.SAMPLE_LLM,
        maxTokens,
      );
      return {
        content: [{ type: "text", text: `LLM sampling result: ${result.content.text}` }],
      };
    }

    if (name === ToolName.GET_TINY_IMAGE) {
      GetTinyImageSchema.parse(args);
      return {
        content: [
          {
            type: "text",
            text: "This is a tiny image:",
          },
          {
            type: "image",
            data: MCP_TINY_IMAGE,
            mimeType: "image/png",
          },
          {
            type: "text",
            text: "The image above is the MCP tiny image.",
          },
        ],
      };
    }

    if (name === ToolName.ANNOTATED_MESSAGE) {
      const { messageType, includeImage } = AnnotatedMessageSchema.parse(args);

      const content = [];

      // Main message with different priorities/audiences based on type
      if (messageType === "error") {
        content.push({
          type: "text",
          text: "Error: Operation failed",
          annotations: {
            priority: 1.0, // Errors are highest priority
            audience: ["user", "assistant"] // Both need to know about errors
          }
        });
      } else if (messageType === "success") {
        content.push({
          type: "text",
          text: "Operation completed successfully",
          annotations: {
            priority: 0.7, // Success messages are important but not critical
            audience: ["user"] // Success mainly for user consumption
          }
        });
      } else if (messageType === "debug") {
        content.push({
          type: "text",
          text: "Debug: Cache hit ratio 0.95, latency 150ms",
          annotations: {
            priority: 0.3, // Debug info is low priority
            audience: ["assistant"] // Technical details for assistant
          }
        });
      }

      // Optional image with its own annotations
      if (includeImage) {
        content.push({
          type: "image",
          data: MCP_TINY_IMAGE,
          mimeType: "image/png",
          annotations: {
            priority: 0.5,
            audience: ["user"] // Images primarily for user visualization
          }
        });
      }

      return { content };
    }

    throw new Error(`Unknown tool: ${name}`);
  });

  server.setRequestHandler(CompleteRequestSchema, async (request) => {
    const { ref, argument } = request.params;

    if (ref.type === "ref/resource") {
      const resourceId = ref.uri.split("/").pop();
      if (!resourceId) return { completion: { values: [] } };

      // Filter resource IDs that start with the input value
      const values = EXAMPLE_COMPLETIONS.resourceId.filter(id =>
        id.startsWith(argument.value)
      );
      return { completion: { values, hasMore: false, total: values.length } };
    }

    if (ref.type === "ref/prompt") {
      // Handle completion for prompt arguments
      const completions = EXAMPLE_COMPLETIONS[argument.name as keyof typeof EXAMPLE_COMPLETIONS];
      if (!completions) return { completion: { values: [] } };

      const values = completions.filter(value =>
        value.startsWith(argument.value)
      );
      return { completion: { values, hasMore: false, total: values.length } };
    }

    throw new Error(`Unknown reference type`);
  });

  server.setRequestHandler(SetLevelRequestSchema, async (request) => {
    const { level } = request.params;
    logLevel = level;

    // Demonstrate different log levels
    await server.notification({
      method: "notifications/message",
      params: {
        level: "debug",
        logger: "test-server",
        data: `Logging level set to: ${logLevel}`,
      },
    });

    return {};
  });

  const cleanup = async () => {
    if (subsUpdateInterval) clearInterval(subsUpdateInterval);
    if (logsUpdateInterval) clearInterval(logsUpdateInterval);
  };

  return { server, cleanup };
};

const MCP_TINY_IMAGE =
  "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";