get_performance_score

Assess website performance by generating a performance score. Input a URL and specify device type (desktop or mobile) to evaluate site efficiency and responsiveness.

Instructions

Get the performance score for a website

Input Schema

TableJSON Schema

Name	Required	Description	Default
`device`	No	Device to emulate (default: desktop)	desktop
`url`	Yes	URL to audit

Implementation Reference

src/tools/performance.ts:42-99 (handler)

MCP tool handler function for 'get_performance_score' that invokes the core getPerformanceScore helper, structures the result, adds recommendations, and handles errors.

async ({ url, device }) => {
  try {
    const result = await getPerformanceScore(url, device);

    const structuredResult = createStructuredPerformance(
      "Performance Score",
      result.url,
      result.device,
      {
        performanceScore: result.performanceScore,
        metrics: Object.fromEntries(
          Object.entries(result.metrics).map(([key, metric]) => [
            key,
            {
              title: metric.title,
              value: metric.displayValue,
              score: metric.score,
            },
          ]),
        ),
        fetchTime: result.fetchTime,
      },
      [
        "Focus on Core Web Vitals improvements",
        "Optimize largest contentful paint for better user experience",
        "Reduce total blocking time to improve interactivity",
      ],
    );

    return {
      content: [
        {
          type: "text" as const,
          text: JSON.stringify(structuredResult, null, 2),
        },
      ],
    };
  } catch (error: unknown) {
    const errorMessage = error instanceof Error ? error.message : String(error);
    return {
      content: [
        {
          type: "text" as const,
          text: JSON.stringify(
            {
              error: "Performance analysis failed",
              url,
              device: device || "desktop",
              message: errorMessage,
            },
            null,
            2,
          ),
        },
      ],
      isError: true,
    };
  }

src/lighthouse-performance.ts:5-14 (handler)

Core handler that runs Lighthouse performance audit, extracts score and metrics.

export async function getPerformanceScore(url: string, device: "desktop" | "mobile" = "desktop") {
  const result = await runLighthouseAudit(url, ["performance"], device);
  return {
    url: result.url,
    device: result.device,
    performanceScore: result.categories.performance?.score || 0,
    metrics: result.metrics,
    fetchTime: result.fetchTime,
  };
}

src/schemas.ts:43-46 (schema)
Input schema definition for the tool, validating url (HTTP/HTTPS) and device (desktop/mobile).
```
export const basicAuditSchema = {
  url: baseSchemas.url,
  device: baseSchemas.device,
};
```

src/tools/performance.ts:38-101 (registration)

Registers the 'get_performance_score' tool on the MCP server with name, description, schema, and handler.

server.tool(
  "get_performance_score",
  "Get the performance score for a website",
  basicAuditSchema,
  async ({ url, device }) => {
    try {
      const result = await getPerformanceScore(url, device);

      const structuredResult = createStructuredPerformance(
        "Performance Score",
        result.url,
        result.device,
        {
          performanceScore: result.performanceScore,
          metrics: Object.fromEntries(
            Object.entries(result.metrics).map(([key, metric]) => [
              key,
              {
                title: metric.title,
                value: metric.displayValue,
                score: metric.score,
              },
            ]),
          ),
          fetchTime: result.fetchTime,
        },
        [
          "Focus on Core Web Vitals improvements",
          "Optimize largest contentful paint for better user experience",
          "Reduce total blocking time to improve interactivity",
        ],
      );

      return {
        content: [
          {
            type: "text" as const,
            text: JSON.stringify(structuredResult, null, 2),
          },
        ],
      };
    } catch (error: unknown) {
      const errorMessage = error instanceof Error ? error.message : String(error);
      return {
        content: [
          {
            type: "text" as const,
            text: JSON.stringify(
              {
                error: "Performance analysis failed",
                url,
                device: device || "desktop",
                message: errorMessage,
              },
              null,
              2,
            ),
          },
        ],
        isError: true,
      };
    }
  },
);

src/index.ts:26-26 (registration)
Top-level call to registerPerformanceTools which includes the get_performance_score tool.
```
registerPerformanceTools(server);
```

Lighthouse MCP