pipeline/packages/shared/src/llm/client.ts

import OpenAI from "openai";

export interface LLMClientConfig {
  baseURL?: string;
  apiKey?: string;
  model: string;
}

export interface LLMResult {
  content: string;
  // Provider finish reason: "stop" (complete), "length" (truncated by
  // max_tokens — yields incomplete JSON), "content_filter", etc. The parse
  // stage uses this to retry truncated github-trending output.
  finishReason: string | null;
}

export class LLMClient {
  private client: OpenAI;
  private model: string;

  constructor(config: LLMClientConfig) {
    this.client = new OpenAI({
      baseURL: config.baseURL || process.env.OPENAI_BASE_URL || "https://api.openai.com/v1",
      apiKey: config.apiKey || process.env.OPENAI_API_KEY,
    });
    this.model = config.model;
  }

  async chat(systemPrompt: string, userMessage: string): Promise<LLMResult> {
    const response = await this.client.chat.completions.create({
      model: this.model,
      messages: [
        { role: "system", content: systemPrompt },
        { role: "user", content: userMessage },
      ],
      temperature: 0.3,
      response_format: { type: "json_object" },
      // github-trending with a large trending list (many repos × verbose
      // intro/narration) can exceed 8k output tokens and get truncated mid-JSON,
      // breaking parsing. 16384 gives headroom. (The gateway clamps rather than
      // errors on an oversized value, and the parse stage additionally retries
      // once with a conciseness nudge if finish_reason="length".)
      max_tokens: 16384,
    });

    const choice = response.choices[0];
    const content = choice?.message?.content;
    if (!content) throw new Error("LLM returned empty response");
    return { content, finishReason: choice?.finish_reason ?? null };
  }
}