> For the complete documentation index, see [llms.txt](https://fater.gitbook.io/fater-ai-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://fater.gitbook.io/fater-ai-docs/core-concepts-ai-and-editor-features/text-prompts.md).

# Text Prompts

Text prompts are your primary way of communicating your desired outcome to Fater's AI models. Whether generating images, videos, or potentially influencing 3D assets, a well-crafted prompt is essential for achieving high-quality, relevant results. A few models also support [Negative Prompts](/fater-ai-docs/core-concepts-ai-and-editor-features/negative-prompts.md) to steer the AI away from unwanted elements.

***

### Positive Prompts (The "What")

* **Purpose:** The main text description telling the AI **what you want to create or see**.
* **Content:** Describe the subject, scene, style, colors, lighting, composition, and any specific details you envision.
* **Where to Use:**
  * The main input field in the **Floating Prompt Area** (for Image, Asset, and Video generation).
  * The **Agent Chat** in the Settings Tab (to instruct the Agent).
  * Specific "Prompt" fields within certain AI Model control panels (though often linked to the Floating Prompt Area).

**Tips for Effective Positive Prompts:**

1. **Be Specific & Detailed:** Instead of `car`, try `red 1960s sports car driving on a coastal road at sunset, cinematic lighting`. More detail gives the AI better direction.
2. **Use Strong Keywords:** Focus on nouns, adjectives, and verbs that clearly define the core elements and style.
3. **Consider Style & Medium:** Add terms like `photorealistic`, `illustration`, `watercolor style`, `cinematic lighting`, `3D render`, `architectural sketch`, `product shot on white background`.
4. **Mention Composition:** Use terms like `wide angle shot`, `close-up portrait`, `top-down view`, `rule of thirds`.
5. **Quality Boosters:** Often helpful to include terms like `masterpiece`, `best quality`, `high resolution`, `detailed`, `sharp focus`. (Some models include these by default).
6. **Iterate:** Your first prompt might not be perfect. Analyze the result and refine your prompt based on what you want to change.