mirror of
https://github.com/modelcontextprotocol/servers.git
synced 2026-02-19 11:54:58 -05:00
Refactor everything.ts to use McpServer.registerTool() for all tools
- Migrated all 13 tools from CallToolRequestSchema handler to individual registerTool() calls - Removed CallToolRequestSchema handler block entirely - Removed unused imports: CallToolRequestSchema, ListToolsRequestSchema, Tool, ToolSchema, zodToJsonSchema - Tools now use Zod schemas directly without zodToJsonSchema conversion - Conditional tools (ELICITATION, LIST_ROOTS) are registered but initially disabled - Tools are enabled in oninitialized based on client capabilities (elicitation, roots) - Fixed progressToken access: changed from extra.progressToken to extra._meta?.progressToken - Added 10-minute timeout to elicitation requests - Removed unused ToolInputSchema and ToolOutputSchema types - All tools maintain their original functionality and logic Migrated tools: 1. ECHO - simple echo tool 2. ADD - adds two numbers 3. LONG_RUNNING_OPERATION - with progress notifications 4. PRINT_ENV - environment variables 5. SAMPLE_LLM - LLM sampling via sendRequest 6. GET_TINY_IMAGE - returns MCP logo image 7. ANNOTATED_MESSAGE - content with annotations 8. GET_RESOURCE_REFERENCE - resource reference content 9. GET_RESOURCE_LINKS - multiple resource links 10. STRUCTURED_CONTENT - with outputSchema 11. ZIP_RESOURCES - creates zip from URLs 12. ELICITATION - user elicitation (conditional) 13. LIST_ROOTS - lists MCP roots (conditional) 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <noreply@anthropic.com>
This commit is contained in:
Olivier Chafik
@@ -1,7 +1,6 @@
|
||||
import { Server } from "@modelcontextprotocol/sdk/server/index.js";
|
||||
import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js";
|
||||
import type { RequestHandlerExtra } from "@modelcontextprotocol/sdk/shared/protocol.js";
|
||||
import {
|
||||
CallToolRequestSchema,
|
||||
ClientCapabilities,
|
||||
CompleteRequestSchema,
|
||||
CreateMessageRequest,
|
||||
@@ -12,7 +11,6 @@ import {
|
||||
ListPromptsRequestSchema,
|
||||
ListResourcesRequestSchema,
|
||||
ListResourceTemplatesRequestSchema,
|
||||
ListToolsRequestSchema,
|
||||
LoggingLevel,
|
||||
ReadResourceRequestSchema,
|
||||
Resource,
|
||||
@@ -20,13 +18,10 @@ import {
|
||||
ServerNotification,
|
||||
ServerRequest,
|
||||
SubscribeRequestSchema,
|
||||
Tool,
|
||||
ToolSchema,
|
||||
UnsubscribeRequestSchema,
|
||||
type Root
|
||||
} from "@modelcontextprotocol/sdk/types.js";
|
||||
import { z } from "zod";
|
||||
import { zodToJsonSchema } from "zod-to-json-schema";
|
||||
import { readFileSync } from "fs";
|
||||
import { fileURLToPath } from "url";
|
||||
import { dirname, join } from "path";
|
||||
@@ -36,12 +31,6 @@ const __filename = fileURLToPath(import.meta.url);
|
||||
const __dirname = dirname(__filename);
|
||||
const instructions = readFileSync(join(__dirname, "instructions.md"), "utf-8");
|
||||
|
||||
const ToolInputSchema = ToolSchema.shape.inputSchema;
|
||||
type ToolInput = z.infer<typeof ToolInputSchema>;
|
||||
|
||||
const ToolOutputSchema = ToolSchema.shape.outputSchema;
|
||||
type ToolOutput = z.infer<typeof ToolOutputSchema>;
|
||||
|
||||
type SendRequest = RequestHandlerExtra<ServerRequest, ServerNotification>["sendRequest"];
|
||||
|
||||
/* Input schemas for tools implemented in this server */
|
||||
@@ -164,10 +153,9 @@ const EXAMPLE_COMPLETIONS = {
|
||||
};
|
||||
|
||||
export const createServer = () => {
|
||||
const server = new Server(
|
||||
const mcpServer = new McpServer(
|
||||
{
|
||||
name: "example-servers/everything",
|
||||
title: "Everything Example Server",
|
||||
version: "1.0.0",
|
||||
},
|
||||
{
|
||||
@@ -182,6 +170,9 @@ export const createServer = () => {
|
||||
}
|
||||
);
|
||||
|
||||
// Access underlying Server for low-level operations
|
||||
const server = mcpServer.server;
|
||||
|
||||
let subscriptions: Set<string> = new Set();
|
||||
let subsUpdateInterval: NodeJS.Timeout | undefined;
|
||||
let stdErrUpdateInterval: NodeJS.Timeout | undefined;
|
||||
@@ -272,7 +263,7 @@ export const createServer = () => {
|
||||
},
|
||||
};
|
||||
|
||||
return await sendRequest(request, ElicitResultSchema);
|
||||
return await sendRequest(request, ElicitResultSchema, {timeout: 10 * 60 * 1000 /* 10 minutes */});
|
||||
};
|
||||
|
||||
const ALL_RESOURCES: Resource[] = Array.from({ length: 100 }, (_, i) => {
|
||||
@@ -479,462 +470,424 @@ export const createServer = () => {
|
||||
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.LONG_RUNNING_OPERATION,
|
||||
description:
|
||||
"Demonstrates a long running operation with progress updates",
|
||||
inputSchema: zodToJsonSchema(LongRunningOperationSchema) as ToolInput,
|
||||
},
|
||||
{
|
||||
name: ToolName.PRINT_ENV,
|
||||
description:
|
||||
"Prints all environment variables, helpful for debugging MCP server configuration",
|
||||
inputSchema: zodToJsonSchema(PrintEnvSchema) 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,
|
||||
},
|
||||
{
|
||||
name: ToolName.GET_RESOURCE_REFERENCE,
|
||||
description:
|
||||
"Returns a resource reference that can be used by MCP clients",
|
||||
inputSchema: zodToJsonSchema(GetResourceReferenceSchema) as ToolInput,
|
||||
},
|
||||
{
|
||||
name: ToolName.GET_RESOURCE_LINKS,
|
||||
description:
|
||||
"Returns multiple resource links that reference different types of resources",
|
||||
inputSchema: zodToJsonSchema(GetResourceLinksSchema) as ToolInput,
|
||||
},
|
||||
{
|
||||
name: ToolName.STRUCTURED_CONTENT,
|
||||
description:
|
||||
"Returns structured content along with an output schema for client data validation",
|
||||
inputSchema: zodToJsonSchema(StructuredContentSchema.input) as ToolInput,
|
||||
outputSchema: zodToJsonSchema(StructuredContentSchema.output) as ToolOutput,
|
||||
},
|
||||
{
|
||||
name: ToolName.ZIP_RESOURCES,
|
||||
description: "Compresses the provided resource files (mapping of name to URI, which can be a data URI) to a zip file, which it returns as a data URI resource link.",
|
||||
inputSchema: zodToJsonSchema(ZipResourcesInputSchema) as ToolInput,
|
||||
}
|
||||
];
|
||||
if (clientCapabilities!.roots) tools.push ({
|
||||
name: ToolName.LIST_ROOTS,
|
||||
description:
|
||||
"Lists the current MCP roots provided by the client. Demonstrates the roots protocol capability even though this server doesn't access files.",
|
||||
inputSchema: zodToJsonSchema(ListRootsSchema) as ToolInput,
|
||||
});
|
||||
if (clientCapabilities!.elicitation) tools.push ({
|
||||
name: ToolName.ELICITATION,
|
||||
description: "Demonstrates the Elicitation feature by asking the user to provide information about their favorite color, number, and pets.",
|
||||
inputSchema: zodToJsonSchema(ElicitationSchema) as ToolInput,
|
||||
});
|
||||
|
||||
return { tools };
|
||||
// Register tools with McpServer
|
||||
mcpServer.registerTool(ToolName.ECHO, {
|
||||
description: "Echoes back the input",
|
||||
inputSchema: {
|
||||
message: z.string().describe("Message to echo")
|
||||
}
|
||||
}, async ({ message }) => {
|
||||
return {
|
||||
content: [{ type: "text", text: `Echo: ${message}` }],
|
||||
};
|
||||
});
|
||||
|
||||
server.setRequestHandler(CallToolRequestSchema, async (request,extra) => {
|
||||
const { name, arguments: args } = request.params;
|
||||
|
||||
if (name === ToolName.ECHO) {
|
||||
const validatedArgs = EchoSchema.parse(args);
|
||||
return {
|
||||
content: [{ type: "text", text: `Echo: ${validatedArgs.message}` }],
|
||||
};
|
||||
mcpServer.registerTool(ToolName.ADD, {
|
||||
description: "Adds two numbers",
|
||||
inputSchema: {
|
||||
a: z.number().describe("First number"),
|
||||
b: z.number().describe("Second number"),
|
||||
}
|
||||
}, async ({ a, b }) => {
|
||||
const sum = a + b;
|
||||
return {
|
||||
content: [{
|
||||
type: "text",
|
||||
text: `The sum of ${a} and ${b} is ${sum}.`,
|
||||
}],
|
||||
};
|
||||
});
|
||||
|
||||
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}.`,
|
||||
},
|
||||
],
|
||||
};
|
||||
mcpServer.registerTool(ToolName.LONG_RUNNING_OPERATION, {
|
||||
description: "Demonstrates a long running operation with progress updates",
|
||||
inputSchema: {
|
||||
duration: z.number().default(10).describe("Duration of the operation in seconds"),
|
||||
steps: z.number().default(5).describe("Number of steps in the operation"),
|
||||
}
|
||||
}, async ({ duration, steps }, extra) => {
|
||||
const stepDuration = duration / steps;
|
||||
const progressToken = extra._meta?.progressToken;
|
||||
|
||||
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,
|
||||
},
|
||||
},{relatedRequestId: extra.requestId});
|
||||
}
|
||||
}
|
||||
|
||||
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,
|
||||
extra.sendRequest
|
||||
for (let i = 1; i < steps + 1; i++) {
|
||||
await new Promise((resolve) =>
|
||||
setTimeout(resolve, stepDuration * 1000)
|
||||
);
|
||||
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:",
|
||||
if (progressToken !== undefined) {
|
||||
await server.notification({
|
||||
method: "notifications/progress",
|
||||
params: {
|
||||
progress: i,
|
||||
total: steps,
|
||||
progressToken,
|
||||
},
|
||||
{
|
||||
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
|
||||
},
|
||||
});
|
||||
}, {relatedRequestId: extra.requestId});
|
||||
}
|
||||
}
|
||||
|
||||
// Optional image with its own annotations
|
||||
if (includeImage) {
|
||||
content.push({
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Long running operation completed. Duration: ${duration} seconds, Steps: ${steps}.`,
|
||||
},
|
||||
],
|
||||
};
|
||||
});
|
||||
|
||||
mcpServer.registerTool(ToolName.PRINT_ENV, {
|
||||
description: "Returns the environment variables",
|
||||
inputSchema: {}
|
||||
}, async () => {
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: JSON.stringify(process.env, null, 2),
|
||||
},
|
||||
],
|
||||
};
|
||||
});
|
||||
|
||||
mcpServer.registerTool(ToolName.SAMPLE_LLM, {
|
||||
description: "Requests a sample from an LLM via the client",
|
||||
inputSchema: {
|
||||
prompt: z.string().describe("The prompt to send to the LLM"),
|
||||
maxTokens: z.number().default(100).describe("Maximum number of tokens to generate"),
|
||||
}
|
||||
}, async ({ prompt, maxTokens }, extra) => {
|
||||
const result = await requestSampling(
|
||||
prompt,
|
||||
ToolName.SAMPLE_LLM,
|
||||
maxTokens,
|
||||
extra.sendRequest
|
||||
);
|
||||
return {
|
||||
content: [
|
||||
{ type: "text", text: `LLM sampling result: ${result.content.text}` },
|
||||
],
|
||||
};
|
||||
});
|
||||
|
||||
mcpServer.registerTool(ToolName.GET_TINY_IMAGE, {
|
||||
description: "Returns a tiny MCP logo image",
|
||||
inputSchema: {}
|
||||
}, async () => {
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: "This is a tiny image:",
|
||||
},
|
||||
{
|
||||
type: "image",
|
||||
data: MCP_TINY_IMAGE,
|
||||
mimeType: "image/png",
|
||||
annotations: {
|
||||
priority: 0.5,
|
||||
audience: ["user"], // Images primarily for user visualization
|
||||
},
|
||||
});
|
||||
}
|
||||
|
||||
return { content };
|
||||
}
|
||||
|
||||
if (name === ToolName.GET_RESOURCE_REFERENCE) {
|
||||
const validatedArgs = GetResourceReferenceSchema.parse(args);
|
||||
const resourceId = validatedArgs.resourceId;
|
||||
|
||||
const resourceIndex = resourceId - 1;
|
||||
if (resourceIndex < 0 || resourceIndex >= ALL_RESOURCES.length) {
|
||||
throw new Error(`Resource with ID ${resourceId} does not exist`);
|
||||
}
|
||||
|
||||
const resource = ALL_RESOURCES[resourceIndex];
|
||||
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Returning resource reference for Resource ${resourceId}:`,
|
||||
},
|
||||
{
|
||||
type: "resource",
|
||||
resource: resource,
|
||||
},
|
||||
{
|
||||
type: "text",
|
||||
text: `You can access this resource using the URI: ${resource.uri}`,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
if (name === ToolName.ELICITATION) {
|
||||
ElicitationSchema.parse(args);
|
||||
|
||||
const elicitationResult = await requestElicitation(
|
||||
'What are your favorite things?',
|
||||
{
|
||||
type: 'object',
|
||||
properties: {
|
||||
color: { type: 'string', description: 'Favorite color' },
|
||||
number: {
|
||||
type: 'integer',
|
||||
description: 'Favorite number',
|
||||
minimum: 1,
|
||||
maximum: 100,
|
||||
},
|
||||
pets: {
|
||||
type: 'string',
|
||||
enum: ['cats', 'dogs', 'birds', 'fish', 'reptiles'],
|
||||
description: 'Favorite pets',
|
||||
},
|
||||
},
|
||||
},
|
||||
extra.sendRequest
|
||||
);
|
||||
|
||||
// Handle different response actions
|
||||
const content = [];
|
||||
|
||||
if (elicitationResult.action === 'accept' && elicitationResult.content) {
|
||||
content.push({
|
||||
{
|
||||
type: "text",
|
||||
text: `✅ User provided their favorite things!`,
|
||||
});
|
||||
text: "The image above is the MCP tiny image.",
|
||||
},
|
||||
],
|
||||
};
|
||||
});
|
||||
|
||||
// Only access elicitationResult.content when action is accept
|
||||
const { color, number, pets } = elicitationResult.content;
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `Their favorites are:\n- Color: ${color || 'not specified'}\n- Number: ${number || 'not specified'}\n- Pets: ${pets || 'not specified'}`,
|
||||
});
|
||||
} else if (elicitationResult.action === 'decline') {
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `❌ User declined to provide their favorite things.`,
|
||||
});
|
||||
} else if (elicitationResult.action === 'cancel') {
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `⚠️ User cancelled the elicitation dialog.`,
|
||||
});
|
||||
}
|
||||
mcpServer.registerTool(ToolName.ANNOTATED_MESSAGE, {
|
||||
description: "Demonstrates content annotations with priority and audience",
|
||||
inputSchema: {
|
||||
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"),
|
||||
}
|
||||
}, async ({ messageType, includeImage }) => {
|
||||
const content: any[] = [];
|
||||
|
||||
// Include raw result for debugging
|
||||
// Main message with different priorities/audiences based on type
|
||||
if (messageType === "error") {
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `\nRaw result: ${JSON.stringify(elicitationResult, null, 2)}`,
|
||||
text: "Error: Operation failed",
|
||||
annotations: {
|
||||
priority: 1.0, // Errors are highest priority
|
||||
audience: ["user", "assistant"], // Both need to know about errors
|
||||
},
|
||||
});
|
||||
|
||||
return { content };
|
||||
}
|
||||
|
||||
if (name === ToolName.GET_RESOURCE_LINKS) {
|
||||
const { count } = GetResourceLinksSchema.parse(args);
|
||||
const content = [];
|
||||
|
||||
// Add intro text
|
||||
} else if (messageType === "success") {
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `Here are ${count} resource links to resources available in this server (see full output in tool response if your client does not support resource_link yet):`,
|
||||
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
|
||||
},
|
||||
});
|
||||
|
||||
// Return resource links to actual resources from ALL_RESOURCES
|
||||
const actualCount = Math.min(count, ALL_RESOURCES.length);
|
||||
for (let i = 0; i < actualCount; i++) {
|
||||
const resource = ALL_RESOURCES[i];
|
||||
content.push({
|
||||
type: "resource_link",
|
||||
uri: resource.uri,
|
||||
name: resource.name,
|
||||
description: `Resource ${i + 1}: ${resource.mimeType === "text/plain"
|
||||
? "plaintext resource"
|
||||
: "binary blob resource"
|
||||
}`,
|
||||
mimeType: resource.mimeType,
|
||||
});
|
||||
}
|
||||
|
||||
return { content };
|
||||
}
|
||||
|
||||
if (name === ToolName.STRUCTURED_CONTENT) {
|
||||
// The same response is returned for every input.
|
||||
const validatedArgs = StructuredContentSchema.input.parse(args);
|
||||
// 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
|
||||
},
|
||||
});
|
||||
}
|
||||
|
||||
const weather = {
|
||||
temperature: 22.5,
|
||||
conditions: "Partly cloudy",
|
||||
humidity: 65
|
||||
}
|
||||
return { content };
|
||||
});
|
||||
|
||||
const backwardCompatiblecontent = {
|
||||
mcpServer.registerTool(ToolName.GET_RESOURCE_REFERENCE, {
|
||||
description: "Returns a resource reference to a static resource",
|
||||
inputSchema: {
|
||||
resourceId: z.number().min(1).max(100).describe("ID of the resource to reference (1-100)"),
|
||||
}
|
||||
}, async ({ resourceId }) => {
|
||||
const resourceIndex = resourceId - 1;
|
||||
if (resourceIndex < 0 || resourceIndex >= ALL_RESOURCES.length) {
|
||||
throw new Error(`Resource with ID ${resourceId} does not exist`);
|
||||
}
|
||||
|
||||
const resource = ALL_RESOURCES[resourceIndex];
|
||||
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Returning resource reference for Resource ${resourceId}:`,
|
||||
},
|
||||
{
|
||||
type: "resource",
|
||||
resource: resource as any,
|
||||
},
|
||||
{
|
||||
type: "text",
|
||||
text: `You can access this resource using the URI: ${resource.uri}`,
|
||||
},
|
||||
] as any,
|
||||
};
|
||||
});
|
||||
|
||||
mcpServer.registerTool(ToolName.GET_RESOURCE_LINKS, {
|
||||
description: "Returns resource links to static resources",
|
||||
inputSchema: {
|
||||
count: z.number().min(1).max(10).default(3).describe("Number of resource links to return (1-10)"),
|
||||
}
|
||||
}, async ({ count }) => {
|
||||
const content: any[] = [];
|
||||
|
||||
// Add intro text
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `Here are ${count} resource links to resources available in this server (see full output in tool response if your client does not support resource_link yet):`,
|
||||
});
|
||||
|
||||
// Return resource links to actual resources from ALL_RESOURCES
|
||||
const actualCount = Math.min(count, ALL_RESOURCES.length);
|
||||
for (let i = 0; i < actualCount; i++) {
|
||||
const resource = ALL_RESOURCES[i];
|
||||
content.push({
|
||||
type: "resource_link",
|
||||
uri: resource.uri,
|
||||
name: resource.name,
|
||||
description: `Resource ${i + 1}: ${resource.mimeType === "text/plain"
|
||||
? "plaintext resource"
|
||||
: "binary blob resource"
|
||||
}`,
|
||||
mimeType: resource.mimeType,
|
||||
});
|
||||
}
|
||||
|
||||
return { content };
|
||||
});
|
||||
|
||||
mcpServer.registerTool(ToolName.STRUCTURED_CONTENT, {
|
||||
description: "Returns structured weather data with both text and typed output",
|
||||
inputSchema: {
|
||||
location: z.string().trim().min(1).describe("City name or zip code"),
|
||||
},
|
||||
outputSchema: {
|
||||
temperature: z.number().describe("Temperature in celsius"),
|
||||
conditions: z.string().describe("Weather conditions description"),
|
||||
humidity: z.number().describe("Humidity percentage"),
|
||||
}
|
||||
}, async ({ location }) => {
|
||||
// The same response is returned for every input.
|
||||
const weather = {
|
||||
temperature: 22.5,
|
||||
conditions: "Partly cloudy",
|
||||
humidity: 65
|
||||
};
|
||||
|
||||
return {
|
||||
content: [{
|
||||
type: "text",
|
||||
text: JSON.stringify(weather)
|
||||
}
|
||||
}],
|
||||
structuredContent: weather
|
||||
};
|
||||
});
|
||||
|
||||
return {
|
||||
content: [backwardCompatiblecontent],
|
||||
structuredContent: weather
|
||||
};
|
||||
mcpServer.registerTool(ToolName.ZIP_RESOURCES, {
|
||||
description: "Creates a zip file from multiple URLs",
|
||||
inputSchema: {
|
||||
files: z.record(z.string().url().describe("URL of the file to include in the zip")).describe("Mapping of file names to URLs to include in the zip"),
|
||||
}
|
||||
}, async ({ files }) => {
|
||||
const zip = new JSZip();
|
||||
|
||||
if (name === ToolName.ZIP_RESOURCES) {
|
||||
const { files } = ZipResourcesInputSchema.parse(args);
|
||||
|
||||
const zip = new JSZip();
|
||||
|
||||
for (const [fileName, fileUrl] of Object.entries(files)) {
|
||||
try {
|
||||
const response = await fetch(fileUrl);
|
||||
if (!response.ok) {
|
||||
throw new Error(`Failed to fetch ${fileUrl}: ${response.statusText}`);
|
||||
}
|
||||
const arrayBuffer = await response.arrayBuffer();
|
||||
zip.file(fileName, arrayBuffer);
|
||||
} catch (error) {
|
||||
throw new Error(`Error fetching file ${fileUrl}: ${error instanceof Error ? error.message : String(error)}`);
|
||||
for (const [fileName, fileUrl] of Object.entries(files)) {
|
||||
try {
|
||||
const response = await fetch(fileUrl);
|
||||
if (!response.ok) {
|
||||
throw new Error(`Failed to fetch ${fileUrl}: ${response.statusText}`);
|
||||
}
|
||||
const arrayBuffer = await response.arrayBuffer();
|
||||
zip.file(fileName, arrayBuffer);
|
||||
} catch (error) {
|
||||
throw new Error(`Error fetching file ${fileUrl}: ${error instanceof Error ? error.message : String(error)}`);
|
||||
}
|
||||
|
||||
const uri = `data:application/zip;base64,${await zip.generateAsync({ type: "base64" })}`;
|
||||
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "resource_link",
|
||||
mimeType: "application/zip",
|
||||
uri,
|
||||
},
|
||||
],
|
||||
};
|
||||
}
|
||||
|
||||
if (name === ToolName.LIST_ROOTS) {
|
||||
ListRootsSchema.parse(args);
|
||||
const uri = `data:application/zip;base64,${await zip.generateAsync({ type: "base64" })}`;
|
||||
|
||||
if (!clientSupportsRoots) {
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: "The MCP client does not support the roots protocol.\n\n" +
|
||||
"This means the server cannot access information about the client's workspace directories or file system roots."
|
||||
}
|
||||
]
|
||||
};
|
||||
}
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "resource_link",
|
||||
uri,
|
||||
name: "resources.zip",
|
||||
mimeType: "application/zip",
|
||||
},
|
||||
],
|
||||
};
|
||||
});
|
||||
|
||||
if (currentRoots.length === 0) {
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: "The client supports roots but no roots are currently configured.\n\n" +
|
||||
"This could mean:\n" +
|
||||
"1. The client hasn't provided any roots yet\n" +
|
||||
"2. The client provided an empty roots list\n" +
|
||||
"3. The roots configuration is still being loaded"
|
||||
}
|
||||
]
|
||||
};
|
||||
}
|
||||
// Conditional tools - store references to enable/disable them based on client capabilities
|
||||
const elicitationTool = mcpServer.registerTool(ToolName.ELICITATION, {
|
||||
description: "Demonstrates user elicitation to collect structured data",
|
||||
inputSchema: {}
|
||||
}, async (args, extra) => {
|
||||
const elicitationResult = await requestElicitation(
|
||||
'What are your favorite things?',
|
||||
{
|
||||
type: 'object',
|
||||
properties: {
|
||||
color: { type: 'string', description: 'Favorite color' },
|
||||
number: {
|
||||
type: 'integer',
|
||||
description: 'Favorite number',
|
||||
minimum: 1,
|
||||
maximum: 100,
|
||||
},
|
||||
pets: {
|
||||
type: 'string',
|
||||
enum: ['cats', 'dogs', 'birds', 'fish', 'reptiles'],
|
||||
description: 'Favorite pets',
|
||||
},
|
||||
},
|
||||
},
|
||||
extra.sendRequest
|
||||
);
|
||||
|
||||
const rootsList = currentRoots.map((root, index) => {
|
||||
return `${index + 1}. ${root.name || 'Unnamed Root'}\n URI: ${root.uri}`;
|
||||
}).join('\n\n');
|
||||
// Handle different response actions
|
||||
const content: any[] = [];
|
||||
|
||||
if (elicitationResult.action === 'accept' && elicitationResult.content) {
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `✅ User provided their favorite things!`,
|
||||
});
|
||||
|
||||
// Only access elicitationResult.content when action is accept
|
||||
const { color, number, pets } = elicitationResult.content;
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `Their favorites are:\n- Color: ${color || 'not specified'}\n- Number: ${number || 'not specified'}\n- Pets: ${pets || 'not specified'}`,
|
||||
});
|
||||
} else if (elicitationResult.action === 'decline') {
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `❌ User declined to provide their favorite things.`,
|
||||
});
|
||||
} else if (elicitationResult.action === 'cancel') {
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `⚠️ User cancelled the elicitation dialog.`,
|
||||
});
|
||||
}
|
||||
|
||||
// Include raw result for debugging
|
||||
content.push({
|
||||
type: "text",
|
||||
text: `\nRaw result: ${JSON.stringify(elicitationResult, null, 2)}`,
|
||||
});
|
||||
|
||||
return { content };
|
||||
});
|
||||
|
||||
const listRootsTool = mcpServer.registerTool(ToolName.LIST_ROOTS, {
|
||||
description: "Lists the current MCP roots provided by the client",
|
||||
inputSchema: {}
|
||||
}, async () => {
|
||||
if (!clientSupportsRoots) {
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Current MCP Roots (${currentRoots.length} total):\n\n${rootsList}\n\n` +
|
||||
"Note: This server demonstrates the roots protocol capability but doesn't actually access files. " +
|
||||
"The roots are provided by the MCP client and can be used by servers that need file system access."
|
||||
text: "The MCP client does not support the roots protocol.\n\n" +
|
||||
"This means the server cannot access information about the client's workspace directories or file system roots."
|
||||
}
|
||||
]
|
||||
};
|
||||
}
|
||||
|
||||
throw new Error(`Unknown tool: ${name}`);
|
||||
if (currentRoots.length === 0) {
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: "The client supports roots but no roots are currently configured.\n\n" +
|
||||
"This could mean:\n" +
|
||||
"1. The client hasn't provided any roots yet\n" +
|
||||
"2. The client provided an empty roots list\n" +
|
||||
"3. The roots configuration is still being loaded"
|
||||
}
|
||||
]
|
||||
};
|
||||
}
|
||||
|
||||
const rootsList = currentRoots.map((root, index) => {
|
||||
return `${index + 1}. ${root.name || 'Unnamed Root'}\n URI: ${root.uri}`;
|
||||
}).join('\n\n');
|
||||
|
||||
return {
|
||||
content: [
|
||||
{
|
||||
type: "text",
|
||||
text: `Current MCP Roots (${currentRoots.length} total):\n\n${rootsList}\n\n` +
|
||||
"Note: This server demonstrates the roots protocol capability but doesn't actually access files. " +
|
||||
"The roots are provided by the MCP client and can be used by servers that need file system access."
|
||||
}
|
||||
]
|
||||
};
|
||||
});
|
||||
|
||||
// Disable conditional tools initially - they'll be enabled in oninitialized
|
||||
elicitationTool.disable();
|
||||
listRootsTool.disable();
|
||||
|
||||
|
||||
server.setRequestHandler(CompleteRequestSchema, async (request) => {
|
||||
const { ref, argument } = request.params;
|
||||
|
||||
@@ -988,12 +941,29 @@ export const createServer = () => {
|
||||
}
|
||||
});
|
||||
|
||||
// Handle post-initialization setup for roots
|
||||
// Handle post-initialization setup for roots and conditional tools
|
||||
server.oninitialized = async () => {
|
||||
clientCapabilities = server.getClientCapabilities();
|
||||
|
||||
// Enable conditional tools based on client capabilities
|
||||
if (clientCapabilities?.experimental?.elicitation) {
|
||||
elicitationTool.enable();
|
||||
await server.sendLoggingMessage({
|
||||
level: "info",
|
||||
logger: "everything-server",
|
||||
data: "Client supports elicitation - enabling ELICITATION tool",
|
||||
}, sessionId);
|
||||
}
|
||||
|
||||
if (clientCapabilities?.roots) {
|
||||
clientSupportsRoots = true;
|
||||
listRootsTool.enable();
|
||||
await server.sendLoggingMessage({
|
||||
level: "info",
|
||||
logger: "everything-server",
|
||||
data: "Client supports roots - enabling LIST_ROOTS tool",
|
||||
}, sessionId);
|
||||
|
||||
try {
|
||||
const response = await server.listRoots();
|
||||
if (response && 'roots' in response) {
|
||||
|
||||
9
src/everything/run-stdio.sh
Executable file
9
src/everything/run-stdio.sh
Executable file
@@ -0,0 +1,9 @@
|
||||
#!/bin/bash
|
||||
# npx -y @modelcontextprotocol/inspector /Users/ochafik/code/modelcontextprotocol-servers/src/everything/run-stdio.sh
|
||||
# files: {"README": "https://raw.githubusercontent.com/modelcontextprotocol/servers/refs/heads/main/README.md"}
|
||||
set -euo pipefail
|
||||
|
||||
cd "$(dirname "$0")"
|
||||
npm i >&2
|
||||
npm run build >&2
|
||||
npm run --silent start
|
||||
Reference in New Issue
Block a user