@studious-lms/server 1.1.12 → 1.1.13

package/src/utils/inference.ts

@@ -0,0 +1,193 @@
+import OpenAI from 'openai';
+import { logger } from './logger.js';
+import { prisma } from '../lib/prisma.js';
+import { pusher } from '../lib/pusher.js';
+import { ensureAIUserExists, getAIUserId } from './aiUser.js';
+
+// Initialize inference client (Cohere via OpenAI SDK)
+
+logger.info('Inference API Key', { apiKey: process.env.INFERENCE_API_KEY });
+logger.info('Inference API Base URL', { baseURL: process.env.INFERENCE_API_BASE_URL });
+
+export const inferenceClient = new OpenAI({
+  apiKey: process.env.INFERENCE_API_KEY,
+  baseURL: process.env.INFERENCE_API_BASE_URL,
+});
+
+// Types for lab chat context
+export interface LabChatContext {
+  subject: string;
+  topic: string;
+  difficulty: 'beginner' | 'intermediate' | 'advanced';
+  objectives: string[];
+  resources?: string[];
+  persona: string;
+  constraints: string[];
+  examples?: any[];
+  metadata?: Record<string, any>;
+}
+
+export interface InferenceResponse {
+  content: string;
+  model: string;
+  tokensUsed: number;
+  finishReason: string;
+}
+
+/**
+ * Centralized function to send AI messages to conversations
+ * Handles database storage and Pusher broadcasting
+ */
+export async function sendAIMessage(
+  content: string,
+  conversationId: string,
+  options: {
+    subject?: string;
+    customSender?: {
+      displayName: string;
+      profilePicture?: string | null;
+    };
+  } = {}
+): Promise<{
+  id: string;
+  content: string;
+  senderId: string;
+  conversationId: string;
+  createdAt: Date;
+}> {
+  // Ensure AI user exists
+  await ensureAIUserExists();
+
+  // Create message in database
+  const aiMessage = await prisma.message.create({
+    data: {
+      content,
+      senderId: getAIUserId(),
+      conversationId,
+    },
+  });
+
+  logger.info('AI Message sent', {
+    messageId: aiMessage.id,
+    conversationId,
+    contentLength: content.length,
+  });
+
+  // Prepare sender info
+  const senderInfo = {
+    id: getAIUserId(),
+    username: 'AI Assistant',
+    profile: {
+      displayName: options.customSender?.displayName || `${options.subject || 'AI'} Assistant`,
+      profilePicture: options.customSender?.profilePicture || null,
+    },
+  };
+
+  // Broadcast via Pusher
+  try {
+    await pusher.trigger(`conversation-${conversationId}`, 'new-message', {
+      id: aiMessage.id,
+      content: aiMessage.content,
+      senderId: getAIUserId(),
+      conversationId: aiMessage.conversationId,
+      createdAt: aiMessage.createdAt,
+      sender: senderInfo,
+      mentionedUserIds: [],
+    });
+  } catch (error) {
+    logger.error('Failed to broadcast AI message:', { error, messageId: aiMessage.id });
+  }
+
+  return {
+    id: aiMessage.id,
+    content: aiMessage.content,
+    senderId: getAIUserId(),
+    conversationId: aiMessage.conversationId,
+    createdAt: aiMessage.createdAt,
+  };
+}
+
+/**
+ * Simple inference function for general use
+ */
+export async function generateInferenceResponse(
+  subject: string,
+  question: string,
+  options: {
+    model?: string;
+    maxTokens?: number;
+  } = {}
+): Promise<InferenceResponse> {
+  const { model = 'command-r-plus', maxTokens = 500 } = options;
+
+  try {
+    const completion = await inferenceClient.chat.completions.create({
+      model,
+      messages: [
+        {
+          role: 'system',
+          content: `You are a helpful educational assistant for ${subject}. Provide clear, concise, and accurate answers. Keep responses educational and appropriate for students.`,
+        },
+        {
+          role: 'user',
+          content: question,
+        },
+      ],
+      max_tokens: maxTokens,
+      temperature: 0.5,
+      // Remove OpenAI-specific parameters for Cohere compatibility
+    });
+
+    const response = completion.choices[0]?.message?.content;
+    
+    if (!response) {
+      throw new Error('No response generated from inference API');
+    }
+
+    return {
+      content: response,
+      model,
+      tokensUsed: completion.usage?.total_tokens || 0,
+      finishReason: completion.choices[0]?.finish_reason || 'unknown',
+    };
+
+  } catch (error) {
+    logger.error('Failed to generate inference response', { error, subject, question: question.substring(0, 50) + '...' });
+    throw error;
+  }
+}
+
+/**
+ * Validate inference configuration
+ */
+export function validateInferenceConfig(): boolean {
+  if (!process.env.INFERENCE_API_KEY) {
+    logger.error('Inference API key not configured for Cohere');
+    return false;
+  }
+  return true;
+}
+
+/**
+ * Get available inference models (for admin/config purposes)
+ */
+export async function getAvailableModels(): Promise<string[]> {
+  try {
+    const models = await inferenceClient.models.list();
+    return models.data
+      .filter(model => model.id.includes('command'))
+      .map(model => model.id)
+      .sort();
+  } catch (error) {
+    logger.error('Failed to fetch inference models', { error });
+    return ['command-r-plus', 'command-r', 'command-light']; // Fallback Cohere models
+  }
+}
+
+/**
+ * Estimate token count for a message (rough approximation)
+ */
+export function estimateTokenCount(text: string): number {
+  // Rough approximation: 1 token ≈ 4 characters for English text
+  return Math.ceil(text.length / 4);
+}