tokencost-mcp-server

Instructions

Implementation Reference

• src/index.ts:90-105 (handler)

  The async handler function for tokencost_get_model_pricing that executes the tool logic. It uses findModel() to look up a model by ID/name, handles not-found cases with suggestions via findModels(), and returns model pricing in both markdown and JSON formats.

  async ({ model }) => {
    const found = findModel(model);
    if (!found) {
      const similar = findModels(model);
      const suggestion = similar.length > 0
        ? `\n\nDid you mean: ${similar.slice(0, 5).map(m => m.id).join(", ")}?`
        : `\n\nAvailable providers: ${providers.join(", ")}. Use tokencost_list_models to see all models.`;
      return {
        content: [{ type: "text", text: `Model "${model}" not found.${suggestion}` }],
      };
    }
    return {
      content: [{ type: "text", text: modelToMarkdown(found) }],
      structuredContent: modelToJSON(found),
    };
  }

• src/index.ts:63-106 (registration)

  The server.registerTool() call that registers the tokencost_get_model_pricing tool with MCP server. Includes tool name, title, description, inputSchema using Zod, annotations, and the handler function.

  server.registerTool(
    "tokencost_get_model_pricing",
    {
      title: "Get Model Pricing",
      description: `Get pricing details for a specific LLM model.
  
  Args:
    - model (string): Model ID or name to look up (e.g., "gpt-5", "claude-sonnet-4.6", "gemini-3-pro")
  
  Returns:
    Model pricing details including input/output costs per 1M tokens, context window, and max output.
    Returns an error message if the model is not found, with suggestions for similar models.
  
  Examples:
    - "gpt-5" → GPT-5 pricing from OpenAI
    - "claude-opus-4.6" → Claude Opus 4.6 pricing from Anthropic
    - "gemini" → First matching Gemini model`,
      inputSchema: {
        model: z.string().min(1).describe("Model ID or name (e.g., 'gpt-5', 'claude-sonnet-4.6')"),
      },
      annotations: {
        readOnlyHint: true,
        destructiveHint: false,
        idempotentHint: true,
        openWorldHint: false,
      },
    },
    async ({ model }) => {
      const found = findModel(model);
      if (!found) {
        const similar = findModels(model);
        const suggestion = similar.length > 0
          ? `\n\nDid you mean: ${similar.slice(0, 5).map(m => m.id).join(", ")}?`
          : `\n\nAvailable providers: ${providers.join(", ")}. Use tokencost_list_models to see all models.`;
        return {
          content: [{ type: "text", text: `Model "${model}" not found.${suggestion}` }],
        };
      }
      return {
        content: [{ type: "text", text: modelToMarkdown(found) }],
        structuredContent: modelToJSON(found),
      };
    }
  );

• src/index.ts:80-82 (schema)

  The Zod input schema for tokencost_get_model_pricing, defining a single 'model' parameter as a non-empty string with a description.

  inputSchema: {
    model: z.string().min(1).describe("Model ID or name (e.g., 'gpt-5', 'claude-sonnet-4.6')"),
  },

• src/pricing.ts:1-10 (schema)

  The ModelPricing TypeScript interface that defines the structure of model pricing data, including id, name, provider, inputPer1M, outputPer1M, contextWindow, maxOutput, and optional notes.

  export interface ModelPricing {
    id: string;
    name: string;
    provider: string;
    inputPer1M: number;
    outputPer1M: number;
    contextWindow: number;
    maxOutput: number;
    notes?: string;
  }

• src/pricing.ts:106-117 (helper)

  The findModel() and findModels() helper functions used by the handler to search for models by ID/name. findModel returns the first match, findModels returns all matching models.

  export function findModel(query: string): ModelPricing | undefined {
    const q = query.toLowerCase();
    return models.find(m => m.id === q || m.name.toLowerCase() === q) ??
      models.find(m => m.id.includes(q) || m.name.toLowerCase().includes(q));
  }
  
  export function findModels(query: string): ModelPricing[] {
    const q = query.toLowerCase();
    return models.filter(m =>
      m.id.includes(q) || m.name.toLowerCase().includes(q) || m.provider.toLowerCase().includes(q)
    );
  }