Skip to main content

​
AI Chat

The AI chat interface is where you build your PRD through conversation. The AI guides you with intelligent questions, extracting product requirements as you talk.

​
The Chat Interface

The chat page has three main areas:
  1. Chat messages — Left/center panel with the conversation
  2. Requirements panel — Right panel showing extracted requirements and completeness score
  3. Input area — Bottom bar for typing, uploading files, or recording voice

​
Starting a Conversation

New project template selection
When you create a project from a template, the AI opens with the template’s initial question. For blank projects, it asks a general question about what you want to build.

​
Suggestion Chips

When the chat is empty, quick-start suggestions appear:
  • “I want to build a web application”
  • “I’m planning a mobile app”
  • “I need a SaaS platform”
Follow-up suggestions appear mid-conversation to help you provide structured answers.

​
Input Methods

​
Text Input

Type your responses in the input field. Press Enter to send or Cmd+Enter for a new line.

​
Document Upload

Upload existing documents for the AI to analyze and incorporate:
  • Supported formats: PDF, DOCX, TXT, Markdown
  • Max size: 10MB
  • How it works: The AI extracts key requirements, target users, technical considerations, and suggested focus areas from your document
Drag and drop files onto the chat area or click the upload button.

​
Voice Input

Record audio for hands-free requirement gathering:
  1. Click the microphone button
  2. Speak your requirements (up to 5 minutes)
  3. Click stop — audio is transcribed via speech-to-text
  4. Transcribed text appears in the chat input for review before sending
Voice input requires the audio transcription feature to be enabled on your plan.

​
Requirements Tracking

As you chat, the AI extracts and displays:
  • Project name and description
  • Problem statement
  • Target users and personas
  • Core features with priority levels
  • Completeness score (0–100%)
The completeness score updates in real time. When it’s high enough, you’ll see a “Ready to Generate PRD” indicator.

​
Message Editing

Made a mistake or want to refine an answer? Click the edit icon on any of your messages to revise it. The AI will re-process the conversation from that point.

​
Generating the PRD

When you’re ready:
  1. Click Generate PRD (appears when completeness is sufficient)
  2. Choose a PRD style:
    • Comprehensive — Full 12-section PRD
    • Technical — Developer-focused specification
    • Business — Stakeholder and ROI focused
    • Lean — MVP abbreviated format
    • Agile — Sprint and iteration focused
  3. The PRD streams in section by section with a progress bar
  4. You’re redirected to the Review page when complete

​
AI Memory

The system automatically extracts reusable insights from your conversations and saves them to your Knowledge Base. Over time, the AI gets better at understanding your preferences, tech stack, and product patterns.

​
Usage Limits

If you hit your plan’s token limit during a conversation, a dialog appears with options to:
  • Upgrade your plan for more tokens
  • Add your own API key (BYOK) for unlimited usage