get_model_pricing

Instructions

Implementation Reference

• mcp-server/src/index.ts:156-186 (registration)

  Registration of the 'get_model_pricing' tool via server.tool() on the MCP server. No input schema. Fetches pricing data from /models endpoint and formats it as text.

  server.tool(
    'get_model_pricing',
    'Get AI model pricing comparison across all major providers (Anthropic, OpenAI, Google, Meta, Mistral, Cohere). Prices per 1M tokens.',
    {},
    async () => {
      const data = await fetchJSON('/models') as {
        providers: {
          name: string;
          models: { name: string; inputPrice: number; outputPrice: number; contextWindow: number; released: string; capabilities: string[] }[];
        }[];
      };
  
      const text = data.providers
        .map(p => {
          const models = p.models
            .map(m => {
              const input = m.inputPrice === 0 ? 'Free' : `$${m.inputPrice.toFixed(2)}`;
              const output = m.outputPrice === 0 ? 'Free' : `$${m.outputPrice.toFixed(2)}`;
              const ctx = m.contextWindow >= 1000000
                ? `${(m.contextWindow / 1000000).toFixed(0)}M`
                : `${(m.contextWindow / 1000).toFixed(0)}K`;
              return `    ${m.name}: Input ${input}, Output ${output}, Context ${ctx}, Released ${m.released}`;
            })
            .join('\n');
          return `  ${p.name}:\n${models}`;
        })
        .join('\n\n');
  
      return { content: [{ type: 'text' as const, text: `AI Model Pricing (per 1M tokens):\n\n${text}` }] };
    }
  );

• mcp-server/src/index.ts:160-185 (handler)

  Handler function for get_model_pricing. Calls fetchJSON('/models') to retrieve pricing data from tensorfeed.ai, then formats it into a human-readable text response listing providers and their models with input/output prices per 1M tokens, context window size, and release date.

  async () => {
    const data = await fetchJSON('/models') as {
      providers: {
        name: string;
        models: { name: string; inputPrice: number; outputPrice: number; contextWindow: number; released: string; capabilities: string[] }[];
      }[];
    };
  
    const text = data.providers
      .map(p => {
        const models = p.models
          .map(m => {
            const input = m.inputPrice === 0 ? 'Free' : `$${m.inputPrice.toFixed(2)}`;
            const output = m.outputPrice === 0 ? 'Free' : `$${m.outputPrice.toFixed(2)}`;
            const ctx = m.contextWindow >= 1000000
              ? `${(m.contextWindow / 1000000).toFixed(0)}M`
              : `${(m.contextWindow / 1000).toFixed(0)}K`;
            return `    ${m.name}: Input ${input}, Output ${output}, Context ${ctx}, Released ${m.released}`;
          })
          .join('\n');
        return `  ${p.name}:\n${models}`;
      })
      .join('\n\n');
  
    return { content: [{ type: 'text' as const, text: `AI Model Pricing (per 1M tokens):\n\n${text}` }] };
  }

• mcp-server/src/index.ts:19-59 (helper)

  Helper function fetchJSON that handles API calls to tensorfeed.ai. Used by the get_model_pricing handler to fetch data from the /models endpoint. Handles authentication via TENSORFEED_TOKEN env var and error states including 402 (payment required) and 401 (token rejected).

  async function fetchJSON(path: string, opts: FetchOptions = {}): Promise<unknown> {
    const headers: Record<string, string> = {
      'User-Agent': `TensorFeed-MCP/${SDK_VERSION}`,
    };
    if (opts.body !== undefined) headers['Content-Type'] = 'application/json';
    if (opts.auth) {
      const token = process.env.TENSORFEED_TOKEN;
      if (!token) {
        throw new Error(
          'TENSORFEED_TOKEN env var is not set. Premium MCP tools require a bearer token. ' +
            'Buy credits at https://tensorfeed.ai/developers/agent-payments and pass the returned tf_live_... token via the TENSORFEED_TOKEN env var in your MCP client config.',
        );
      }
      headers['Authorization'] = `Bearer ${token}`;
    }
    const res = await fetch(`${API_BASE}${path}`, {
      method: opts.method ?? 'GET',
      headers,
      ...(opts.body !== undefined ? { body: JSON.stringify(opts.body) } : {}),
    });
    if (!res.ok) {
      let errPayload: unknown;
      try {
        errPayload = await res.json();
      } catch {
        errPayload = await res.text().catch(() => '');
      }
      if (res.status === 402) {
        throw new Error(
          `Payment required (402). Your token may be out of credits. Top up at https://tensorfeed.ai/developers/agent-payments. Detail: ${JSON.stringify(errPayload)}`,
        );
      }
      if (res.status === 401) {
        throw new Error(
          `Token rejected (401). Check that TENSORFEED_TOKEN is set to a valid tf_live_... token. Detail: ${JSON.stringify(errPayload)}`,
        );
      }
      throw new Error(`API error ${res.status}: ${JSON.stringify(errPayload)}`);
    }
    return res.json();
  }