Rag powerd assitant

.claude/settings.local.json

@@ -0,0 +1,9 @@
+{
+  "permissions": {
+    "allow": [
+      "Bash(cat:*)"
+    ],
+    "deny": [],
+    "ask": []
+  }
+}

package.json

@@ -13,6 +13,8 @@
     "setup-admin": "tsx scripts/setup-admin.ts"
   },
   "dependencies": {
+    "@google/generative-ai": "^0.24.1",
+    "@pinecone-database/pinecone": "^6.1.2",
     "@react-three/drei": "^10.7.6",
     "@react-three/fiber": "^9.3.0",
     "@types/three": "^0.180.0",

scripts/check-api-keys.ts

@@ -0,0 +1,52 @@
+/**
+ * Check API Keys Configuration
+ * Verifies that Gemini and Pinecone API keys are properly loaded
+ */
+
+console.log('🔑 Checking API Keys Configuration\n');
+console.log('='.repeat(60));
+
+// Check Gemini API Key
+const geminiKey = process.env.GOOGLE_GEMINI_API_KEY;
+console.log('\n📊 Gemini API Key:');
+if (geminiKey) {
+  console.log(`✅ Configured (${geminiKey.substring(0, 20)}...)`);
+} else {
+  console.log('❌ NOT configured');
+  console.log('⚠️  Set GOOGLE_GEMINI_API_KEY in .env.local');
+}
+
+// Check Pinecone API Key
+const pineconeKey = process.env.PINECONE_API_KEY;
+console.log('\n📊 Pinecone API Key:');
+if (pineconeKey) {
+  console.log(`✅ Configured (${pineconeKey.substring(0, 20)}...)`);
+} else {
+  console.log('❌ NOT configured');
+  console.log('⚠️  Set PINECONE_API_KEY in .env.local');
+}
+
+// Check MongoDB URI
+const mongoUri = process.env.MONGODB_URI;
+console.log('\n📊 MongoDB URI:');
+if (mongoUri) {
+  console.log('✅ Configured');
+} else {
+  console.log('❌ NOT configured');
+}
+
+console.log('\n' + '='.repeat(60));
+
+if (geminiKey && pineconeKey) {
+  console.log('\n✅ All API keys are configured!');
+  console.log('\n📝 Next steps:');
+  console.log('1. If you just added/changed keys, restart your Next.js dev server');
+  console.log('2. Run: npm run dev');
+  console.log('3. Test the chat assistant in your browser');
+} else {
+  console.log('\n❌ Some API keys are missing!');
+  console.log('\n📝 Fix:');
+  console.log('1. Check that .env.local exists in project root');
+  console.log('2. Add missing keys to .env.local');
+  console.log('3. Restart Next.js dev server: npm run dev');
+}

scripts/check-api-rate-limit.ts

@@ -0,0 +1,156 @@
+/**
+ * Check Gemini API Rate Limit Status
+ * This script tests your API key and shows current limits
+ */
+
+import dotenv from 'dotenv';
+import path from 'path';
+
+// Load environment variables
+dotenv.config({ path: path.join(process.cwd(), '.env.local') });
+dotenv.config({ path: path.join(process.cwd(), '.env') });
+
+import { GoogleGenerativeAI } from '@google/generative-ai';
+
+async function checkRateLimit() {
+  console.log('🔍 Checking Gemini API Rate Limit Status...\n');
+  console.log('=' .repeat(70));
+
+  // Check if API key exists
+  const apiKey = process.env.GOOGLE_GEMINI_API_KEY;
+  if (!apiKey) {
+    console.log('❌ ERROR: GOOGLE_GEMINI_API_KEY not found in .env.local');
+    console.log('   Please add your API key to .env.local');
+    return;
+  }
+
+  console.log('✅ API Key found:', apiKey.substring(0, 10) + '...' + apiKey.substring(apiKey.length - 4));
+  console.log('');
+
+  // Test different models
+  const models = [
+    'gemini-2.0-flash-exp',      // Current model (10 RPM)
+    'gemini-1.5-flash',          // Alternative (15 RPM)
+    'gemini-1.5-pro'             // Pro version (2 RPM)
+  ];
+
+  const genAI = new GoogleGenerativeAI(apiKey);
+
+  console.log('📊 Testing API access with different models:\n');
+
+  for (const modelName of models) {
+    try {
+      console.log(`Testing: ${modelName}...`);
+
+      const model = genAI.getGenerativeModel({ model: modelName });
+      const startTime = Date.now();
+
+      const result = await model.generateContent('Say "Hello" in one word');
+      const response = result.response.text();
+
+      const endTime = Date.now();
+      const duration = endTime - startTime;
+
+      console.log(`  ✅ SUCCESS`);
+      console.log(`  Response: "${response.trim()}"`);
+      console.log(`  Latency: ${duration}ms`);
+      console.log('');
+
+    } catch (error) {
+      const err = error as { status?: number; statusText?: string; message?: string; errorDetails?: Array<{ '@type': string; violations?: Array<{ quotaMetric: string; quotaValue: string }>; retryDelay?: string }> };
+
+      if (err.status === 429) {
+        console.log(`  ❌ RATE LIMIT EXCEEDED`);
+        console.log(`  Status: ${err.status} ${err.statusText}`);
+
+        // Extract rate limit info from error
+        if (err.errorDetails) {
+          const quotaFailure = err.errorDetails.find((d) =>
+            d['@type'] === 'type.googleapis.com/google.rpc.QuotaFailure'
+          );
+
+          if (quotaFailure?.violations) {
+            quotaFailure.violations.forEach((v) => {
+              console.log(`  Quota Metric: ${v.quotaMetric}`);
+              console.log(`  Quota Value: ${v.quotaValue} requests per minute`);
+            });
+          }
+
+          const retryInfo = err.errorDetails.find((d) =>
+            d['@type'] === 'type.googleapis.com/google.rpc.RetryInfo'
+          );
+
+          if (retryInfo?.retryDelay) {
+            console.log(`  Retry After: ${retryInfo.retryDelay}`);
+          }
+        }
+        console.log('');
+      } else if (err.status === 404) {
+        console.log(`  ⚠️  MODEL NOT AVAILABLE (may require different tier)`);
+        console.log('');
+      } else {
+        console.log(`  ❌ ERROR: ${err.message}`);
+        console.log('');
+      }
+    }
+
+    // Wait between tests to avoid rate limit
+    await new Promise(resolve => setTimeout(resolve, 1000));
+  }
+
+  console.log('=' .repeat(70));
+  console.log('\n📋 Rate Limit Information:\n');
+
+  console.log('Free Tier Limits (per minute):');
+  console.log('  • gemini-2.0-flash-exp:  10 RPM (current model)');
+  console.log('  • gemini-1.5-flash:      15 RPM (recommended alternative)');
+  console.log('  • gemini-1.5-pro:         2 RPM (more powerful but slower)');
+  console.log('');
+
+  console.log('💡 Tips:');
+  console.log('  1. For testing: Add 6-second delays between requests');
+  console.log('  2. For production: Current limits are fine (users space out questions)');
+  console.log('  3. If you hit limits: System falls back to keyword search automatically');
+  console.log('');
+
+  // Test rapid requests to see limit
+  console.log('🧪 Testing rapid requests to find your actual limit...\n');
+  console.log('Sending 5 quick requests in succession...');
+
+  const testModel = genAI.getGenerativeModel({ model: 'gemini-2.0-flash-exp' });
+  let successCount = 0;
+  let failCount = 0;
+
+  for (let i = 1; i <= 5; i++) {
+    try {
+      const result = await testModel.generateContent(`Count: ${i}`);
+      console.log(`  ${i}. ✅ Success - ${result.response.text().trim()}`);
+      successCount++;
+      await new Promise(resolve => setTimeout(resolve, 500));
+    } catch (error) {
+      const err = error as { status?: number };
+      if (err.status === 429) {
+        console.log(`  ${i}. ❌ Rate Limited - Hit quota at request #${i}`);
+        failCount++;
+        break;
+      }
+    }
+  }
+
+  console.log('');
+  console.log('📊 Rapid Test Results:');
+  console.log(`  Successful: ${successCount}/5`);
+  console.log(`  Rate Limited: ${failCount > 0 ? 'Yes' : 'No'}`);
+
+  if (failCount > 0) {
+    console.log(`  \n  ⚠️  You're currently at or near your rate limit.`);
+    console.log(`     Wait ~60 seconds before running more tests.`);
+  } else {
+    console.log(`  \n  ✅ You have quota available for more requests.`);
+  }
+
+  console.log('\n' + '=' .repeat(70));
+  console.log('\n✅ Rate limit check complete!');
+}
+
+checkRateLimit().catch(console.error);

scripts/debug-joke.ts

@@ -0,0 +1,11 @@
+import { isSpamOrOffTopic } from '../src/lib/geminiService';
+
+const message = 'tell me a joke';
+const result = isSpamOrOffTopic(message);
+
+console.log(`Message: "${message}"`);
+console.log(`Is spam/off-topic: ${result}`);
+console.log(`Expected: true`);
+
+// Check if 'joke' is in the message
+console.log(`\nContains 'joke': ${message.toLowerCase().includes('joke')}`);