This article explains how to successfully implement a Universal or Custom Document Agent.
Prerequisites
|
Understanding Document Agents
Document Agents allow users to generate presentations with AI. Using the chat interface, users can explain what they want to create, and the agents will generate presentations that are on brand, compliant, and using the company's tone of voice. Compliance is ensured by defining the layout template that will be used for presentation and slide generation.
Admin vs. User Control
| Admin | End-User | |
|---|---|---|
| Slide Layouts | ✔ (Determine available layouts) | ✔ (Edit chosen layouts during presentation creation) |
| Tone of Voice | ✔ | |
Knowledge |
✔ (Via Pre-configured Context) | ✔ (Via Uploaded Files) |
| Image Library | ✔ |
Optimize your Template
Document Agents will follow the guidance of a defined presentation template. To ensure the AI model uses the layouts and placeholders of this template appropriately, follow the best practices for configuring your slide layouts.
Define a Tone of Voice
Setting the tone of voice centrally ensures that the GenAI outputs stay consistent with your brand and preferred style of writing.
Organize Image Library
Document Agents use images from your Templafy library. To ensure the agents find and use the right images, your library needs to be properly organized:
- Ensure your image library has enough content. If the library is too small, the agent won't have many options to choose from.
- Use descriptive names instead of random filenames.
- Use tags to help the agents search the library more effectively.
- Use the description field to add context that filenames and tags cannot capture (ex. aesthetic, mood, setting).
Comments
Article is closed for comments.