NPM Context Agent MCP

Instructions

Implementation Reference

• src/index.ts:936-997 (handler)

  The inline anonymous async handler function that implements the core logic of the get_package_quality tool. It fetches quality metrics from the npms.io API, validates the response using QualityMetricsSchema, computes formatted text output, and returns structured content with metrics.
  async ({ packageName }) => { try { const encodedPackageName = encodeURIComponent(packageName); const response = await fetch( `https://api.npms.io/v2/package/${encodedPackageName}` ); if (!response.ok) { throw new Error( `Failed to fetch quality metrics: ${response.statusText}` ); } const rawData = await response.json(); const parseResult = QualityMetricsSchema.safeParse(rawData.score); if (!parseResult.success) { throw new Error( `Invalid quality metrics structure: ${parseResult.error.message}` ); } const score = parseResult.data; const output = { name: packageName, final: score.final, quality: score.detail.quality, popularity: score.detail.popularity, maintenance: score.detail.maintenance, }; const formattedText = `Package: ${packageName}\n\nQuality Metrics:\n Overall Score: ${( score.final * 100 ).toFixed(1)}%\n Quality: ${(score.detail.quality * 100).toFixed( 1 )}%\n Popularity: ${(score.detail.popularity * 100).toFixed( 1 )}%\n Maintenance: ${(score.detail.maintenance * 100).toFixed(1)}%`; return { content: [ { type: "text", text: formattedText, }, ], structuredContent: output, }; } catch (error) { return { content: [ { type: "text", text: `Error fetching quality metrics: ${ error instanceof Error ? error.message : "Unknown error" }`, }, ], isError: true, }; } }
• src/index.ts:920-998 (registration)

  The server.registerTool call that registers the get_package_quality tool, specifying its name, title, description, inputSchema (packageName: string), outputSchema (name, final, quality, popularity, maintenance scores), and references the handler function.
  server.registerTool( "get_package_quality", { title: "Get Package Quality Metrics", description: "Get quality metrics from npms.io", inputSchema: { packageName: z.string(), }, outputSchema: { name: z.string(), final: z.number(), quality: z.number(), popularity: z.number(), maintenance: z.number(), }, }, async ({ packageName }) => { try { const encodedPackageName = encodeURIComponent(packageName); const response = await fetch( `https://api.npms.io/v2/package/${encodedPackageName}` ); if (!response.ok) { throw new Error( `Failed to fetch quality metrics: ${response.statusText}` ); } const rawData = await response.json(); const parseResult = QualityMetricsSchema.safeParse(rawData.score); if (!parseResult.success) { throw new Error( `Invalid quality metrics structure: ${parseResult.error.message}` ); } const score = parseResult.data; const output = { name: packageName, final: score.final, quality: score.detail.quality, popularity: score.detail.popularity, maintenance: score.detail.maintenance, }; const formattedText = `Package: ${packageName}\n\nQuality Metrics:\n Overall Score: ${( score.final * 100 ).toFixed(1)}%\n Quality: ${(score.detail.quality * 100).toFixed( 1 )}%\n Popularity: ${(score.detail.popularity * 100).toFixed( 1 )}%\n Maintenance: ${(score.detail.maintenance * 100).toFixed(1)}%`; return { content: [ { type: "text", text: formattedText, }, ], structuredContent: output, }; } catch (error) { return { content: [ { type: "text", text: `Error fetching quality metrics: ${ error instanceof Error ? error.message : "Unknown error" }`, }, ], isError: true, }; } } );
• src/index.ts:115-123 (schema)

  Zod schema definition for validating the 'score' object from npms.io API response, used in the handler's safeParse call.
  const QualityMetricsSchema = z.object({ final: z.number(), detail: z.object({ quality: z.number(), popularity: z.number(), maintenance: z.number(), }), analyzedAt: z.string().optional(), });