> ## Documentation Index
> Fetch the complete documentation index at: https://docs.talkcut.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Quickstart

> Create your first AI-generated image and video in minutes. This guide walks you through the basics of TalkCut.

This guide takes you from sign-up to your first AI-generated video in about five minutes.

<Steps>
  <Step title="Create your workspace">
    Go to [talkcut.com](https://talkcut.com) and sign up with your Google account or email. After signing in, you land on your workspace dashboard — this is where all your projects live.
  </Step>

  <Step title="Start a new project">
    Click **Create project** on the dashboard. Give it a name and you'll enter the editor.

    The editor has two main areas:

    * **Canvas** (left) — your visual workspace where generated content appears
    * **Chat panel** (right) — where you talk to the AI agent
  </Step>

  <Step title="Generate your first image">
    In the chat panel, describe what you want to see. For example:

    ```
    A futuristic cityscape at sunset with flying vehicles and neon signs
    ```

    The AI agent processes your request and places the generated image directly on the canvas. You can drag, resize, and rearrange it.
  </Step>

  <Step title="Turn it into a video">
    Now that you have an image on the canvas, ask the AI agent to animate it:

    ```
    Animate this image — slow camera pan across the cityscape with vehicles moving
    ```

    The agent takes your canvas image and generates a video clip from it. The video appears on your canvas.
  </Step>

  <Step title="Try variations">
    Every generation creates a new item on the canvas — nothing gets overwritten. Try different approaches:

    * Ask for a different style: *"Make this more cinematic with warmer colors"*
    * Edit the image: *"Remove the vehicles and add rain"*
    * Generate from a different angle: *"Same scene but from street level at night"*

    You can compare all variations side by side on the canvas.
  </Step>

  <Step title="Upload reference material">
    For more control, upload your own images using the **Upload** button in the toolbar. Reference images help the AI understand your creative intent — the vision engine analyzes style, composition, and color from your uploads.

    Uploaded files go to your [asset library](/workspace/assets), where they are automatically tagged and organized by visual similarity. You can reuse them across projects.
  </Step>

  <Step title="Use skills for consistency">
    If you find yourself repeating the same style directions, save them as a [skill](/workspace/skills). Go to **Skills** in your workspace sidebar and create a reusable prompt template with a `/slash` command.

    Next time you chat with the AI agent, type `/` to invoke your skill instead of retyping the instructions.
  </Step>
</Steps>

## What to explore next

<Columns cols={2}>
  <Card title="Canvas" icon="objects-column" href="/core-concepts/canvas">
    Learn about shapes, frames, and spatial organization on the canvas.
  </Card>

  <Card title="Generating images" icon="image" href="/creating/generating-images">
    Explore all image generation, editing, and enhancement options.
  </Card>

  <Card title="Generating videos" icon="video" href="/creating/generating-videos">
    Create videos from text, keyframes, and reference images.
  </Card>

  <Card title="Workspace memory" icon="brain" href="/core-concepts/workspace-memory">
    Set up persistent brand context that improves generation over time.
  </Card>
</Columns>

<Tip>
  Each generation uses credits from your plan. Check the [pricing page](/workspace/pricing) for details on credit costs per operation.
</Tip>