pixellab-mcp/dist/tools/estimateSkeleton.js

MCP server for PixelLab pixel art generation and manipulation. Usage: pixellab-mcp --secret=your-key

import { z } from "zod";
import { Base64Image } from "@pixellab-code/pixellab";
import { toMcpResponse, toMcpError } from "../utils.js";
export const estimateSkeletonSchema = z.object({
    image_path: z
        .string()
        .describe("Path to the character image to analyze for skeleton/pose detection"),
    show_image: z
        .boolean()
        .default(false)
        .describe("Whether to show the original image with skeleton data to the AI assistant for viewing and analysis"),
});
export async function estimateSkeleton(args, client) {
    try {
        // Load image
        const image = await Base64Image.fromFile(args.image_path);
        // Direct passthrough to pixellab-js
        const response = await client.estimateSkeleton({
            image,
        });
        const description = `Estimated skeleton with ${response.keypoints.length} keypoints detected`;
        const metadata = {
            parameters: args,
            keypointCount: response.keypoints.length,
            keypoints: response.keypoints.map((kp) => ({
                label: kp.label,
                position: { x: kp.x, y: kp.y },
                zIndex: kp.zIndex,
            })),
            usage: response.usage,
            timestamp: new Date().toISOString(),
        };
        if (args.show_image) {
            // Show original image with skeleton data to AI
            return toMcpResponse(description, image, metadata);
        }
        else {
            // Text-only response with skeleton data
            return {
                content: [
                    { type: "text", text: description },
                    { type: "text", text: `Cost: $${response.usage.usd} USD` },
                    { type: "text", text: JSON.stringify(metadata, null, 2) },
                ],
            };
        }
    }
    catch (error) {
        return toMcpError(error);
    }
}