import { McpServer } from "@modelcontextprotocol/sdk/server/mcp.js"
import { z } from "zod"
// Configuration schema for Maiga API
export const configSchema = z.object({
apiToken: z.string().describe("Partner API token for Maiga API authentication"),
})
// Stateless server (default) - no stateful export needed, but explicitly declared for build compatibility
// Helper function to make API requests
async function makeApiRequest(
baseUrl: string,
apiToken: string,
endpoint: string,
body: any,
debug: boolean = false
): Promise<any> {
const url = `${baseUrl}${endpoint}`
if (debug) {
console.log(`[DEBUG] Making request to: ${url}`)
console.log(`[DEBUG] Request body:`, JSON.stringify(body, null, 2))
}
const response = await fetch(url, {
method: "POST",
headers: {
"X-PARTNER-TOKEN": apiToken,
"Content-Type": "application/json",
},
body: JSON.stringify(body),
})
const responseText = await response.text()
if (debug) {
console.log(`[DEBUG] Response status: ${response.status}`)
console.log(`[DEBUG] Response body:`, responseText)
}
if (!response.ok) {
let errorMessage = `API request failed with status ${response.status}`
try {
const errorData = JSON.parse(responseText)
errorMessage = errorData.message || errorData.error || errorMessage
// Handle rate limiting
if (response.status === 429) {
const retryAfter = errorData.retry_after_seconds || "unknown"
errorMessage = `Rate limit exceeded. ${errorMessage}. Retry after ${retryAfter} seconds.`
}
} catch {
// If parsing fails, use the raw response text
errorMessage = responseText || errorMessage
}
throw new Error(errorMessage)
}
try {
return JSON.parse(responseText)
} catch {
throw new Error(`Failed to parse API response: ${responseText}`)
}
}
export default function createServer({
config,
}: {
config: z.infer<typeof configSchema>
}) {
const server = new McpServer({
name: "Maiga API",
version: "1.0.0",
})
// Base URL for Maiga API
const baseUrl = "https://api.maiga.ai"
// Tool 1: Token Analysis
server.registerTool(
"maiga_analyse_token",
{
title: "Token Analysis",
description: "Performs comprehensive technical and fundamental analysis on a cryptocurrency token",
inputSchema: {
identifier: z.string().describe("Token symbol (e.g., 'bitcoin', 'ethereum', 'BTC') or contract address"),
},
},
async ({ identifier }) => {
try {
const result = await makeApiRequest(
baseUrl,
config.apiToken,
"/partner/analyse",
{ identifier },
)
return {
content: [
{
type: "text",
text: JSON.stringify(result, null, 2),
},
],
}
} catch (error) {
return {
content: [
{
type: "text",
text: `Error analyzing token: ${error instanceof Error ? error.message : String(error)}`,
},
],
}
}
},
)
// Tool 2: Mindshare Analysis
server.registerTool(
"maiga_mindshare",
{
title: "Mindshare Analysis",
description: "Analyzes social media sentiment and trending discussions about a token over the last 24 hours",
inputSchema: {
identifier: z.string().describe("Token symbol or contract address"),
},
},
async ({ identifier }) => {
try {
const result = await makeApiRequest(
baseUrl,
config.apiToken,
"/partner/mindshare",
{ identifier },
)
return {
content: [
{
type: "text",
text: JSON.stringify(result, null, 2),
},
],
}
} catch (error) {
return {
content: [
{
type: "text",
text: `Error analyzing mindshare: ${error instanceof Error ? error.message : String(error)}`,
},
],
}
}
},
)
// Tool 3: Token Information
server.registerTool(
"maiga_token_info",
{
title: "Token Information",
description: "Retrieves detailed token holder information and on-chain analysis",
inputSchema: {
identifier: z.string().describe("Token contract address or identifier"),
},
},
async ({ identifier }) => {
try {
const result = await makeApiRequest(
baseUrl,
config.apiToken,
"/partner/token-info",
{ identifier },
)
return {
content: [
{
type: "text",
text: JSON.stringify(result, null, 2),
},
],
}
} catch (error) {
return {
content: [
{
type: "text",
text: `Error fetching token info: ${error instanceof Error ? error.message : String(error)}`,
},
],
}
}
},
)
// Tool 4: Market Reports
server.registerTool(
"maiga_market_report",
{
title: "Market Reports",
description: "Generates specialized market reports based on different analysis modes",
inputSchema: {
mode: z
.enum(["Market_Behavior", "Open_Interest", "Multi_Timeframe", "Fund_Flow"])
.describe("Analysis mode: Market_Behavior, Open_Interest, Multi_Timeframe, or Fund_Flow"),
},
},
async ({ mode }) => {
try {
const result = await makeApiRequest(
baseUrl,
config.apiToken,
"/partner/report",
{ mode },
)
return {
content: [
{
type: "text",
text: JSON.stringify(result, null, 2),
},
],
}
} catch (error) {
return {
content: [
{
type: "text",
text: `Error generating market report: ${error instanceof Error ? error.message : String(error)}`,
},
],
}
}
},
)
// Tool 5: KOL Analysis
server.registerTool(
"maiga_kol_analysis",
{
title: "KOL Analysis",
description: "Analyzes the influence and statistics of cryptocurrency influencers on X (Twitter)",
inputSchema: {
username: z.string().describe("Twitter username (without @) of the KOL to analyze"),
},
},
async ({ username }) => {
try {
const result = await makeApiRequest(
baseUrl,
config.apiToken,
"/partner/kol",
{ username },
)
return {
content: [
{
type: "text",
text: JSON.stringify(result, null, 2),
},
],
}
} catch (error) {
return {
content: [
{
type: "text",
text: `Error analyzing KOL: ${error instanceof Error ? error.message : String(error)}`,
},
],
}
}
},
)
// Tool 6: Trending Tokens
server.registerTool(
"maiga_trending_tokens",
{
title: "Trending Tokens",
description: "Retrieves the top trending tokens in the last 24 hours based on social media mentions and activity",
inputSchema: {},
},
async () => {
try {
const result = await makeApiRequest(
baseUrl,
config.apiToken,
"/partner/trending-token",
{},
)
return {
content: [
{
type: "text",
text: JSON.stringify(result, null, 2),
},
],
}
} catch (error) {
return {
content: [
{
type: "text",
text: `Error fetching trending tokens: ${error instanceof Error ? error.message : String(error)}`,
},
],
}
}
},
)
return server.server
}
