Skip to main content
Glama
luizzzvictor

mcp-comexstat

by luizzzvictor

queryHistoricalData

Retrieve historical trade data by specifying flow, time period, detailed metrics, and filters for exports or imports using the MCP server.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
detailsYes
filtersNo
flowYes
languageNopt
metricsYes
monthDetailNo
periodYes

Implementation Reference

  • MCP tool handler that receives parameters, calls ComexstatClient.queryHistoricalData, stringifies the result as JSON, and returns it as text content.
    async ({ flow, period, monthDetail, filters, details, metrics, language, }) => ({ content: [ { type: "text", text: JSON.stringify( await this.client.queryHistoricalData( flow, period, monthDetail, filters || [], details, metrics, language ) ), }, ], })
  • Zod input schema for validating parameters of the queryHistoricalData MCP tool, including flow type, date period format, optional filters, required details and metrics arrays.
    { flow: z.enum(["export", "import"]), period: z.object({ from: z .string() .regex( /^\d{4}-\d{2}$/, "Must be in YYYY-MM format (e.g., '2023-01')" ), to: z .string() .regex( /^\d{4}-\d{2}$/, "Must be in YYYY-MM format (e.g., '2023-01')" ), }), monthDetail: z.boolean().optional().default(false), filters: z .array( z.object({ filter: z.string(), values: z.array(z.number()).or(z.array(z.string())), }) ) .optional(), details: z.array(z.string()), metrics: z.array(z.string()), language: z.string().optional().default("pt"), },
  • Registration of the 'queryHistoricalData' MCP tool using server.tool(), specifying the tool name, input schema, and execution handler.
    "queryHistoricalData", { flow: z.enum(["export", "import"]), period: z.object({ from: z .string() .regex( /^\d{4}-\d{2}$/, "Must be in YYYY-MM format (e.g., '2023-01')" ), to: z .string() .regex( /^\d{4}-\d{2}$/, "Must be in YYYY-MM format (e.g., '2023-01')" ), }), monthDetail: z.boolean().optional().default(false), filters: z .array( z.object({ filter: z.string(), values: z.array(z.number()).or(z.array(z.string())), }) ) .optional(), details: z.array(z.string()), metrics: z.array(z.string()), language: z.string().optional().default("pt"), }, async ({ flow, period, monthDetail, filters, details, metrics, language, }) => ({ content: [ { type: "text", text: JSON.stringify( await this.client.queryHistoricalData( flow, period, monthDetail, filters || [], details, metrics, language ) ), }, ], }) );
  • Supporting client method implementation that validates input dates, constructs POST request to '/historical-data' endpoint with parameters, and returns the API response. This is called by the MCP handler.
    async queryHistoricalData( flow: "export" | "import", period: { from: string; to: string }, // Format: YYYY-MM (e.g., "2023-01") monthDetail: boolean = false, filters: Array<{ filter: string; values: number[] | string[] }> = [], details: string[], metrics: string[], language: string = "pt" ): Promise<{ data: Array<{ year: number; monthNumber?: number; metricFOB: string; metricKG: string; [key: string]: any; }>; success: boolean; message: string | null; processo_info: any; language: string; }> { // Validate date format const dateFormatRegex = /^\d{4}-\d{2}$/; if ( !dateFormatRegex.test(period.from) || !dateFormatRegex.test(period.to) ) { throw new Error( 'Period dates must be in YYYY-MM format (e.g., "2023-01")' ); } return this.post( "/historical-data", { flow, monthDetail, period, filters, details, metrics, }, { language } ); }

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/luizzzvictor/mcp-comexstat'

If you have feedback or need assistance with the MCP directory API, please join our Discord server