# Providing Additional Context Using PDFs

Enrich DinoAI's context with project-specific information contained in PDF documentation, specifications, or reports.

Your project might have unique requirements, business rules, or technical specifications documented in PDFs that need to be incorporated into your models. DinoAI can extract and understand information from PDF documents, allowing it to incorporate domain-specific knowledge into its responses.

### Example Prompt

> *Based on the data model described in the attached architecture PDF, create a dimensional model for tracking user engagement*

{% hint style="info" %}
**Requiremen**t: Add relevant PDF documents as [context](https://docs.paradime.io/app-help/documentation/context/file-context#id-3.-file-selection) in DinoAI before sending your prompt.
{% endhint %}

***

### Step-by-Step Instructions

1. **Add the PDF document(s)** to your project 
2. **Open DinoAI** by clicking the DinoAI icon (🪄) in the left panel
3. **Add PDF as** [**context**](https://docs.paradime.io/app-help/documentation/dino-ai/context/file-context) **by** clicking the "@" symbol in the Dino AI chat panel
4. **Formulate your prompt** that references the information in the PDF:
   * Be specific about which parts of the PDF you're interested in
   * Clearly state what you want DinoAI to do with the information
5. **Review DinoAI's response** and implement the solution
6. ***Optional*****: Ask follow-up questions** if needed to refine the understanding

***

### How It Works

After you enter your prompt:

1. DinoAI reads and processes the PDF document you've provided as [context](https://docs.paradime.io/app-help/documentation/dino-ai/context/file-context)
2. It extracts text, tables, and structural information from the document
3. It analyzes this information in relation to your specific request
4. It applies the relevant knowledge from the PDF to your task
5. If configured, DinoAI follows your .dinorules preferences when generating its response

{% hint style="info" %}
**Note**: Text-based PDFs work better than scanned documents. For best results, provide PDFs with clear text content rather than handwritten notes or poor-quality scans.
{% endhint %}

***

### Key Benefits

* **Domain Expertise**: Incorporate business-specific knowledge into your models
* **Consistency**: Ensure models align with documented specifications
* **Documentation Integration**: Bridge the gap between documentation and implementation
* **Knowledge Transfer**: Convert PDF information into actionable code
* **Reduced Context Switching**: No need to reference PDFs separately during development

***

#### Example Use Cases

* **Implementing Documented Requirements**: Create models based on specifications in project documentation
* **Following Defined Business Rules**: Apply business logic described in PDF reports
* **Understanding Data Structures**: Reference data dictionaries or schema documentation
* **Translating Technical Specs**: Convert written requirements into working dbt™ models
* **Analyzing Reports**: Use information from PDF reports to inform analytical model design

{% hint style="info" %}
**Tip**: For recurring reference documents, consider adding instructions in your [.dinorules](https://docs.paradime.io/app-help/documentation/dino-ai/dino-rules) file specifying how DinoAI should interpret and handle information from specific PDFs.
{% endhint %}