Generate AI name
Copy for LLM
Copy page as Markdown for LLMs
View as Markdown
Open this page as Markdown
Open in ChatGPT
Get insights from ChatGPT
Open in Claude
Get insights from Claude

🔨 In Development — This section is still being developed and may change.

Generate an intelligent thread title using AI based on conversation content. Supports streaming responses.

POSThttps://api.freddy.aitronos.com/v1/threads/{thread_id}/generateName

Generate an intelligent thread title using AI based on conversation content. Uses the last 10 messages for context and generates concise, descriptive titles. Supports streaming responses for real-time feedback.

You can optionally provide messages in the request body to avoid waiting for database saves (recommended for real-time UI updates).

Path Parameters

thread_id string required

The unique identifier of the thread to generate a name for.

Query Parameters

stream boolean optional · Defaults to false

Enable streaming mode to receive the generated name incrementally as it's created.

Request Body (Optional)

messages array optional

Array of message objects to use for name generation. If provided, these messages will be used instead of fetching from the database. This allows for instant name generation without waiting for database saves.

role string required - Message role (user or assistant)
content string required - Message content

Behavior:

If messages provided: Uses messages from request body (no database lookup)
If messages not provided: Fetches messages from database (existing behavior)

Bash

curl -X POST "https://api.freddy.aitronos.com/v1/threads/thread_abc123/generateName" \
  -H "Authorization: Bearer $FREDDY_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "messages": [
      {
        "role": "user",
        "content": "How do I configure Docker volumes for my PostgreSQL database?"
      },
      {
        "role": "assistant",
        "content": "To configure Docker volumes for PostgreSQL, you can use named volumes..."
      }
    ]
  }'

Response

{
  "success": true,
  "threadId": "thread_abc123",
  "name": "Customer Support Query",
  "message": "Thread name generated successfully"
}

AI Naming Logic

The AI name generation uses the following logic:

Context Analysis: Analyzes up to 10 messages from the thread (or provided messages)
Title Generation: Creates concise, descriptive titles (2-5 words, maximum 50 characters)
Focus on User Intent: Prioritizes user messages to understand what the user is asking about
Intelligent Cleanup: Removes unnecessary prefixes, quotes, trailing punctuation
Title Case Formatting: Ensures proper capitalization for readability
Fallback Handling: Returns "Untitled" if generation fails
Content Awareness: Considers conversation topics, user intent, and context

Name Generation Model

Uses ftg-3.0-speed for fast, efficient title generation
No default system instructions to prevent bias
Focused prompt explicitly emphasizing user messages and context
Internal usage tracking (not visible in UI)

Use Cases

Recommended: Pass Messages in Request Body

✅ Best for:

Real-time UI updates (no database wait)
Streaming responses (generate name immediately after response)
storeResponses=false scenarios
Avoiding race conditions with message saving

// After streaming response completes
const nameResponse = await fetch(`/v1/threads/${threadId}/generateName`, {
  method: 'POST',
  headers: { 'Authorization': `Bearer ${apiKey}`, 'Content-Type': 'application/json' },
  body: JSON.stringify({
    messages: [
      { role: 'user', content: userMessage },
      { role: 'assistant', content: assistantResponse }
    ]
  })
});

Alternative: Fetch from Database