@restnfeel/agentc-starter-kit/src/rag/summarizer/summarizer.ts

한국어 기업용 CMS 모듈 - Task Master AI와 함께 빠르게 웹사이트를 구현할 수 있는 재사용 가능한 컴포넌트 시스템

import "openai/shims/node";
import OpenAI from "openai";

export interface SummarizerOptions {
  maxChunkLength?: number;
  summaryLength?: number;
  focus?: string;
  dangerouslyAllowBrowser?: boolean;
}

export interface SummarizationResult {
  summary: string;
  fullText: string;
  metadata: Record<string, any>;
}

export class WebpageSummarizer {
  private openai: OpenAI;
  private options: SummarizerOptions;

  constructor(apiKey: string, options: SummarizerOptions = {}) {
    this.openai = new OpenAI({ apiKey, dangerouslyAllowBrowser: true });
    this.options = {
      maxChunkLength: 2000,
      summaryLength: 300,
      ...options,
    };
  }

  public async summarize(
    content: string,
    metadata: Record<string, any> = {}
  ): Promise<SummarizationResult> {
    const chunks = this.chunkContent(content, this.options.maxChunkLength!);
    const summaries = await Promise.all(
      chunks.map((chunk) =>
        this.summarizeChunk(
          chunk,
          this.options.summaryLength,
          this.options.focus
        )
      )
    );
    const summary = summaries.join("\n");
    return {
      summary,
      fullText: content,
      metadata,
    };
  }

  private chunkContent(content: string, maxLen: number): string[] {
    const result: string[] = [];
    let i = 0;
    while (i < content.length) {
      result.push(content.slice(i, i + maxLen));
      i += maxLen;
    }
    return result;
  }

  private async summarizeChunk(
    chunk: string,
    summaryLength = 300,
    focus?: string
  ): Promise<string> {
    const prompt = `다음 웹페이지 본문을 ${summaryLength}자 이내로 핵심만 요약하세요.${
      focus ? "\n요약 초점: " + focus : ""
    }\n\n본문:\n${chunk}`;
    const res = await this.openai.chat.completions.create({
      model: "gpt-3.5-turbo",
      messages: [
        { role: "system", content: "당신은 웹페이지 요약 전문가입니다." },
        { role: "user", content: prompt },
      ],
      max_tokens: summaryLength,
      temperature: 0.3,
    });
    return res.choices[0].message?.content?.trim() || "";
  }
}