Skip to content

feat: Implement conversation filtering to prevent saving trivial and repeated messages to AI training data. #382

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
codeunia-dev merged 1 commit into from
Nov 26, 2025
Merged

feat: Implement conversation filtering to prevent saving trivial and repeated messages to AI training data. #382

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
163 changes: 132 additions & 31 deletions app/api/ai/route.ts
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -82,6 +82,93 @@ function checkRateLimit(ip: string): { allowed: boolean; error?: string } {
return { allowed: true };
}

// Filter function to determine if conversation should be saved
async function shouldSaveConversation(message: string, userId: string): Promise<boolean> {
const trimmedMessage = message.trim().toLowerCase();

// Filter 1: Simple greetings
const simpleGreetings = [
/^hi!*$/,
/^hello!*$/,
/^hey!*$/,
/^hii+!*$/,
/^sup!*$/,
/^yo!*$/,
/^hai!*$/,
/^helo!*$/,
/^hllo!*$/,
];

for (const pattern of simpleGreetings) {
if (pattern.test(trimmedMessage)) {
console.log(`Filtering out simple greeting: "${message}"`);
return false;
}
}

// Filter 2: Casual chit-chat (very short messages or common casual phrases)
const casualPhrases = [
/^thanks?!*$/,
/^thank you!*$/,
/^ok!*$/,
/^okay!*$/,
/^cool!*$/,
/^nice!*$/,
/^lol!*$/,
/^haha+!*$/,
/^yeah!*$/,
/^yep!*$/,
/^nope?!*$/,
/^sure!*$/,
/^k!*$/,
];

for (const pattern of casualPhrases) {
if (pattern.test(trimmedMessage)) {
console.log(`Filtering out casual chit-chat: "${message}"`);
return false;
}
}

// Filter out very short messages (less than 10 characters)
if (trimmedMessage.length < 10) {
console.log(`Filtering out short message: "${message}" (${trimmedMessage.length} chars)`);
return false;
}

// Filter 3: Repeated questions (check if same user asked exact same question in last 24 hours)
try {
const supabase = getSupabaseClient();
const twentyFourHoursAgo = new Date(Date.now() - 24 * 60 * 60 * 1000).toISOString();

const { data, error } = await supabase
.from('ai_training_data')
.select('query_text')
.eq('user_id', userId)
.eq('query_text', message)
.gte('created_at', twentyFourHoursAgo)
.limit(1);

if (error) {
console.error('Error checking for duplicate questions:', error);
// If there's an error, allow saving to be safe
return true;
}

if (data && data.length > 0) {
console.log(`Filtering out repeated question: "${message}"`);
return false;
}
} catch (error) {
console.error('Error in duplicate check:', error);
// If there's an error, allow saving to be safe
return true;
}

// If none of the filters matched, save the conversation
return true;
}

// Configure the runtime for this API route
export const runtime = 'nodejs';

Expand Down Expand Up @@ -1083,20 +1170,27 @@ export async function POST(request: NextRequest) {

// Save to database after stream completes
try {
const sessionId = crypto.randomUUID();
const supabase = getSupabaseClient();

await supabase
.from('ai_training_data')
.insert({
user_id: userId,
session_id: sessionId,
query_text: message,
response_text: fullResponse,
context_type: finalContext
});

console.log(`Streaming conversation saved to database (User: ${userId})`);
// Check if conversation should be saved
const shouldSave = await shouldSaveConversation(message, userId);

if (shouldSave) {
const sessionId = crypto.randomUUID();
const supabase = getSupabaseClient();

await supabase
.from('ai_training_data')
.insert({
user_id: userId,
session_id: sessionId,
query_text: message,
response_text: fullResponse,
context_type: finalContext
});

console.log(`Streaming conversation saved to database (User: ${userId})`);
} else {
console.log(`Streaming conversation filtered out (User: ${userId})`);
}
} catch (dbError) {
console.error('Error saving streaming conversation:', dbError);
}
Expand Down Expand Up @@ -1164,24 +1258,31 @@ export async function POST(request: NextRequest) {

// Save conversation to database for training and analytics
try {
const sessionId = crypto.randomUUID();
const supabase = getSupabaseClient();

const { error: dbError } = await supabase
.from('ai_training_data')
.insert({
user_id: userId || null,
session_id: sessionId,
query_text: message,
response_text: aiResponse,
context_type: finalContext
});

if (dbError) {
console.error('Failed to save AI conversation:', dbError);
// Check if conversation should be saved
const shouldSave = await shouldSaveConversation(message, userId);

if (shouldSave) {
const sessionId = crypto.randomUUID();
const supabase = getSupabaseClient();

const { error: dbError } = await supabase
.from('ai_training_data')
.insert({
user_id: userId || null,
session_id: sessionId,
query_text: message,
response_text: aiResponse,
context_type: finalContext
});

if (dbError) {
console.error('Failed to save AI conversation:', dbError);
} else {
const userInfo = userId ? `(User: ${userId})` : '(Anonymous)';
console.log(`AI conversation saved successfully to database ${userInfo}`);
}
} else {
const userInfo = userId ? `(User: ${userId})` : '(Anonymous)';
console.log(`AI conversation saved successfully to database ${userInfo}`);
console.log(`Non-streaming conversation filtered out (User: ${userId})`);
}
} catch (dbSaveError) {
console.error('Error saving to database:', dbSaveError);
Expand Down
Loading