naisys/dist/features/genimg.js

Node.js Autonomous Intelligence System

import OpenAI from "openai";
import path from "path";
import sharp from "sharp";
import * as config from "../config.js";
import * as costTracker from "../llm/costTracker.js";
import * as output from "../utils/output.js";
import * as pathService from "../utils/pathService.js";
import { NaisysPath } from "../utils/pathService.js";
/** genimg "<description>" <filepath>: Generate an image with the description and save it to the file path */
export async function handleCommand(args) {
    // genimg sholdn't even be presented as an available command unless it is defined in the config
    if (!config.agent.imageModel) {
        throw "Agent config: Error, 'imageModel' is not defined";
    }
    const newParams = args.split('"');
    if (newParams.length < 3) {
        throw "Invalid parameters: Description in quotes and fully qualified filepath with desired image extension are required";
    }
    const description = newParams[1].trim();
    const filepath = new NaisysPath(newParams[2].trim() || "");
    if (!description) {
        throw "Error: Description is required";
    }
    if (!filepath) {
        throw "Error: Filepath is required";
    }
    // Validate path is fully qualified
    if (!filepath.getNaisysPath().startsWith("/")) {
        throw "Error: Filepath must be fully qualified";
    }
    pathService.ensureFileDirExists(filepath);
    output.comment(`Generating image with ${config.agent.imageModel}...`);
    const openai = new OpenAI();
    const model = getImageModel(config.agent.imageModel);
    const response = await openai.images.generate({
        prompt: description,
        model: model.name,
        size: model.size,
        quality: model.quality,
        response_format: "b64_json",
    });
    // save to filepath
    const base64Image = response.data[0].b64_json;
    if (!base64Image) {
        throw 'Error: "b64_json" not found in response';
    }
    // Convert the base64 string to a buffer
    const imageBuffer = Buffer.from(base64Image, "base64");
    // Use sharp to convert the buffer and save it as a JPG file
    const hostPath = filepath.toHostPath();
    const fileExtension = path.extname(hostPath).substring(1);
    await sharp(imageBuffer)
        /*.resize(512, 512, {
          fit: "inside",
        })*/
        .toFormat(fileExtension)
        .toFile(hostPath);
    // Record the cost
    await costTracker.recordCost(model.cost, "genimg", model.name);
    return "1024x1024 Image generated and saved to " + filepath.getNaisysPath();
}
const imageModels = [
    {
        key: "dalle3-1024-HD",
        name: "dall-e-3",
        size: "1024x1024",
        quality: "hd",
        cost: 0.08,
    },
    {
        key: "dalle3-1024",
        name: "dall-e-3",
        size: "1024x1024",
        cost: 0.04,
    },
    {
        key: "dalle2-1024",
        name: "dall-e-2",
        size: "1024x1024",
        cost: 0.02,
    },
    {
        key: "dalle2-512",
        name: "dall-e-2",
        size: "512x512",
        cost: 0.018,
    },
    {
        key: "dalle2-256",
        name: "dall-e-2",
        size: "256x256",
        cost: 0.016,
    },
];
function getImageModel(key) {
    const model = imageModels.find((m) => m.key === key);
    if (!model) {
        throw `Error, image model not found: ${key}`;
    }
    return model;
}
//# sourceMappingURL=genimg.js.map