Automating Power BI Documentation with Microsoft Fabric Data Agents

This presentation explores a transformative approach to Power BI documentation using Microsoft Fabric Data Agents, showcasing how artificial intelligence, automation, and multi-source data integration can replace the traditionally manual and error-prone process of documenting Power BI models.

Manual documentation is one of the most time-consuming and overlooked aspects of BI model management. It often leads to inconsistencies, outdated information, and an over-reliance on a few technical experts. Microsoft Fabric Data Agents change this landscape by automating documentation tasks using advanced capabilities such as natural language processing, semantic understanding, and integration with tools like INFO.CALCDEPENDENCY, SemPy, and Fabric notebooks.

Participants will gain a deep understanding of how to set up and configure Fabric Data Agents to automatically extract metadata, analyze model dependencies, and generate documentation that is both accurate and easy to understand. We will explore how these agents can interpret plain English queries, dynamically generate SQL, DAX, or KQL queries, and return user-friendly responses—democratizing access to analytics and fostering a data-literate culture across teams.

The session will cover the full lifecycle of implementing automated documentation, from setting up Fabric workspaces and enabling the necessary tenant settings, to configuring data access, preparing metadata, and adding natural language instructions to guide the agent. Real-world scenarios will demonstrate how to link agents with Lakehouses, Power BI semantic models, and KQL databases, enabling seamless multi-source connectivity and deeper model insights.

In addition, we will address best practices for data preparation, including structuring, naming conventions, and metadata enrichment, to maximize the effectiveness of AI-generated documentation. You'll also learn how to automate maintenance tasks using notebooks and schedule documentation updates, ensuring that your records remain current and aligned with evolving models.

Another key focus will be on the scalability of this approach. We will examine how to automate documentation across multiple workspaces using PowerShell, REST APIs, and the Fabric Data Agent SDK. Integration with CI/CD pipelines and version control systems like GitHub will also be discussed, offering a code-like management approach to documentation. Export options to Markdown and CSV provide further flexibility, enabling teams to share and archive documentation in formats tailored to their needs.

Finally, we will review security and governance considerations, including role-based access control, data encryption, and audit log monitoring, as well as strategies for troubleshooting, performance tuning, and continuous monitoring of agent health.

By the end of this presentation, attendees will be equipped with the knowledge and tools needed to deploy, manage, and scale intelligent documentation workflows using Microsoft Fabric Data Agents. Whether you're a BI professional, data engineer, or organizational leader, this session will show you how to save time, reduce errors, and gain new insights through intelligent automation—making Power BI documentation smarter, faster, and more accessible than ever before.

Transcript

1. Leveraging AI and Multi-Source Integration for Efficient, Accurate, and Scalable Data Model Documentation Automating Power BI Documentation with Microsoft Fabric Data Agents
2. 03 02 01 Automating Power BI Documentation with Microsoft Fabric Data Agents This automation saves hours of manual work and reduces errors. It makes Power BI models easier to understand and maintain. Benefits of Automation in Power BI Microsoft Fabric Data Agents offer a smart, AI-driven solution to automate explanations and keep documentation current. These agents leverage tools like INFO.CALCDEPENDENCY, SemPy, and Fabric notebooks to handle complex data dependencies. AI-Driven Solutions with Microsoft Fabric Data Agents The Challenge of Manual Documentation Manual Power BI documentation is time-consuming and often overwhelming. This process can lead to outdated and error-prone documentation. Introduction to Automating Power BI Documentation 2
3. Automating Power BI Documentation with Microsoft Fabric Data Agents Key Takeaways 3 01 02 03 04 05 Seamless Documentation Sharing Export documentation to Markdown or CSV for flexibility. Share it seamlessly via Microsoft Teams, SharePoint, or Power BI Apps. Generating Accurate Documentation Clear AI instructions and well- organized data enable accurate documentation. The agent produces easy-to- understand documentation for users. Extracting Model Dependencies Use the INFO.CALCDEPENDENCY DAX function to extract detailed model dependencies. Automate data extraction using SemPy in notebooks for efficiency. Setting Up for Optimal Performance Setup requires enabling tenant settings and securing data access. Preparing your data environment is essential for optimal performance. Automating Power BI Documentation Microsoft Fabric Data Agents use AI to automate Power BI documentation. This reduces manual effort and answers questions in plain English.
4. 01 02 04 03 Automating Power BI Documentation with Microsoft Fabric Data Agents By removing the complexity of data queries, these agents empower all team members to engage with analytics. This democratizes data access and fosters a data-driven culture within organizations. Empowering Teams with Analytics Agents connect you to diverse data sources such as Lakehouses, Power BI datasets, and KQL databases. This ensures a seamless experience when working with various data environments. Seamless Data Connectivity These agents understand natural language queries using large language models (LLMs) powered by Azure OpenAI. No need to write complex queries; simply ask questions in plain English and get precise answers. Natural Language Understanding Intelligent Assistants for Data Microsoft Fabric Data Agents are intelligent assistants for your data environment. They simplify data interaction, making analytics accessible to all team members. What Are Microsoft Fabric Data Agents? 4
5. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents The agent executes the query and returns a clear, easy-to-read answer. This eliminates the need for manual query writing and accelerates data insights. Execution and Response The query is checked for accuracy and security compliance. This ensures the generated query adheres to organizational and technical standards. Accuracy and Security Compliance The agent rewrites your question if necessary and generates a query in SQL, DAX, or KQL. This step transforms your natural language input into a technical query format. Query Generation The agent verifies your permissions and identifies the best data source. It ensures the data source is appropriate, whether it's a Lakehouse, Power BI dataset, or KQL database. Permission and Data Source Identification You type a question or request in natural language. This serves as the starting point for the agent to understand your query. Input in Natural Language How Fabric Data Agents Work 5
6. Automating Power BI Documentation with Microsoft Fabric Data Agents Multi-Source Data Access 6 03 02 01 The integration enables more complete documentation across data sources. It supports advanced analytics for deeper and more actionable insights. Enhanced Documentation and Insights This multi-source capability provides a unified view across all data layers. It spans from raw data to advanced analytics for richer insights. Comprehensive Data Layer Integration Simultaneous Multi-Source Connectivity Microsoft Fabric Data Agents can connect to up to five different data sources at the same time. These sources include Lakehouses, Power BI semantic models, and KQL databases.
7. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents Quick Insight Sharing Insights can be shared rapidly across teams. Avoid sifting through lengthy documents for information. Transparent Explanations Receive clear explanations detailing data usage. Understand how answers are derived with clarity. Contextual Conversations Maintain conversation context for seamless follow-ups. This ensures continuity in data-related discussions. Plain English Interaction Data becomes accessible to all team members. Users can interact with data using simple language. Automate Documentation Generative AI eliminates the need for manual writing. This automation streamlines the documentation process. 7 Why Use Microsoft Fabric Data Agents?
8. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents XMLA Endpoint for Power BI If using Power BI semantic models, the XMLA endpoint must be enabled. Confirm this with your admin to avoid setup delays. Accessible Data Source Ensure at least one accessible data source such as a warehouse, lakehouse, Power BI semantic model, or KQL database. This is necessary for connecting and utilizing data effectively. Cross-Geo AI and Data Storage Enable cross-geo AI processing and data storage for optimal performance. This setting is crucial for handling distributed data efficiently. Tenant Settings Configuration Tenant settings must be enabled for Fabric Data Agents and Copilot. This ensures seamless integration and functionality. Fabric Capacity Requirement To get started with Fabric Data Agents, ensure you have a paid Fabric capacity resource at F2 tier or higher. This is a fundamental requirement for the setup process. Prerequisites for Setup 8
9. Automating Power BI Documentation with Microsoft Fabric Data Agents Security Considerations 9 Use Azure Private Link and Virtual Network connectivity to keep data traffic private. This approach minimizes exposure to public networks and enhances security. Private Network Connectivity Encrypt data both in transit and at rest to safeguard information. Encryption ensures that data remains secure even if intercepted. Data Encryption Practices Monitor audit logs for unusual activity. This practice helps in identifying and responding to potential security threats. Audit Log Monitoring Apply Microsoft Purview sensitivity labels to protect sensitive data. These labels help classify and secure data based on its sensitivity level. Sensitivity Labels for Data Protection Implement role-based access control to restrict agent usage to authorized users. This ensures that only authorized personnel can interact with Fabric Data Agents. Role-Based Access Control 01 02 03 04 05
10. Automating Power BI Documentation with Microsoft Fabric Data Agents Preparing Your Data Environment 10 Implement standards, access controls, and lineage tracking. Maintain data quality and compliance across the environment. Enforcing Data Governance Boost performance with indexes, caching, and materialized views. Ensure data retrieval is fast and reliable for analytics. Tuning Lakehouse and Warehouse Enhance Power BI performance with aggregated tables and Import mode. Clean data models to improve query efficiency and reliability. Optimizing Power BI Queries Create efficient pipelines with batch processing and parallel execution. Monitor for bottlenecks to maintain seamless data flow. Designing Data Factory Pipelines Optimize OneLake storage by partitioning data by time or geography. Use Delta Lake format for transactional data to ensure smooth performance. Organizing OneLake Storage 01 02 03 04 05
11. Automating Power BI Documentation with Microsoft Fabric Data Agents Overcoming Setup Challenges 11 01 02 03 04 05 Frequent reviews of security settings help avoid slowdowns and access problems. This proactive measure safeguards the system's performance and reliability. Prioritizing Security Settings Regular cleaning of Delta tables and monitoring pipelines is essential. This practice prevents data-related issues and ensures smooth operations. Ensuring Data Hygiene Begin with one data source to showcase real-world benefits. This approach helps in building confidence and understanding among stakeholders. Starting Small for Big Impact Teams often struggle to trust new automated tools. Demonstrating their reliability and value is key to overcoming this hurdle. Building Trust in Automation Connecting diverse data sources securely and reliably is a common challenge. This requires robust systems to ensure seamless and secure integration. Challenges in Data Source Integration
12. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents Publish the agent to make it operational. This step finalizes the setup and enables automated documentation. Publishing Your Agent Add natural language instructions describing data meaning and usage. Provide example questions to guide agent responses effectively. Adding Instructions and Examples Select the data source, such as a lakehouse with documentation tables. Pick specific tables the agent can access to tailor its functionality. Configuring Data Access Name your agent clearly to reflect its purpose. This ensures the agent is easily identifiable and aligned with its intended function. Naming and Purpose Definition Open your Fabric workspace and select '+ New Item.' Choose 'Fabric data agent' from the list to begin the setup process. Initiating Your Fabric Workspace Creating Your First Fabric Data Agent 12
13. Automating Power BI Documentation with Microsoft Fabric Data Agents Automate data refreshes and transformations within notebooks for ongoing maintenance. This keeps your data up-to-date and ensures long-term reliability. Automated Maintenance Validate data quality to improve agent interpretation accuracy. High-quality data leads to better and more precise documentation generation. Data Quality Validation Clean and format data to remove inconsistencies or unnecessary brackets. This ensures the data is structured and ready for analysis. Data Cleaning and Formatting Use notebooks to ingest data into your Lakehouse, ensuring completeness. This step is crucial for setting up a reliable foundation for your agent. Data Ingestion Using Notebooks to Prepare Data 13
14. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents Adopting this approach significantly improves scalability and operational efficiency. It allows for streamlined management of agents across various environments. Enhance Scalability and Efficiency Schedule automated documentation runs to maintain up-to-date records. This can be integrated with CI/CD pipelines for continuous deployment. Automate Documentation and Scheduling Create, configure, and update agents programmatically within your workflows. This ensures that agents are tailored to meet specific operational requirements. Programmatic Agent Configuration Utilize Fabric client libraries available in Python or other supported languages. These libraries enable seamless integration and interaction with the system. Leverage Fabric Client Libraries Set up environment variables such as PROJECT_ENDPOINT and FABRIC_CONNECTION_ID. These variables are essential for configuring the agent's operational context. Define Environment Variables Programmatic Agent Management 14
15. Automating Power BI Documentation with Microsoft Fabric Data Agents Enable the agent to read metadata and perform best practice checks without a user account. This ensures secure, automated access to Power BI model metadata for documentation. Step 3: Enable Metadata Access Assign admin permissions to the service principal in Power BI workspaces. Use REST API or PowerShell scripts to complete this step effectively. Step 2: Assign Admin Permissions Securely link your agent to Power BI workspaces by creating an Azure App Registration. This step provides a service principal necessary for secure access. Step 1: Create Azure App Registration Connecting Fabric Data Agents to Power BI 15
16. Automating Power BI Documentation with Microsoft Fabric Data Agents 01 02 03 04 Leverage advanced integration to unlock automated insights and documentation for your Power BI models. This step enhances the overall efficiency and intelligence of your workspace. Advanced Integration Benefits Allow AI clients to query workspace metadata programmatically for deeper insights. This capability supports automated documentation generation for Power BI models. Programmatic Metadata Querying Set up a secure MCP server authenticated with your service principal. This ensures safe and reliable communication between AI clients and your workspace. Run a Secure MCP Server Enable GraphQL introspection on your Fabric workspace to enhance automation capabilities. This step lays the foundation for programmatic access to workspace metadata. Activate GraphQL Introspection Advanced Integration with GraphQL Introspection 16
17. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents This detailed dependency data is essential for accurate documentation and impact analysis. Understanding these dependencies ensures better management and updates to the model. Importance of Dependency Data Analyze the output to understand how calculations relate and depend on each other. This analysis helps in identifying the relationships and dependencies within the model. Analyzing Dependency Data Ensure you have 'Discover' or admin permissions to execute this function. Without the appropriate permissions, the function cannot be run successfully. Permission Requirements Run INFO.CALCDEPENDENCY() to retrieve a table showing dependencies among measures, columns, and tables. This function provides a structured view of how different elements in the model are interrelated. Executing the INFO.CALCDEPENDENCY Function Connect to your Power BI semantic model via Analysis Services or dual Power BI Desktop instances. This connection is the first step to accessing the model's dependencies. Connecting to the Semantic Model Extracting Model Dependencies with INFO.CALCDEPENDENCY 17
18. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents This technique deepens your understanding of model structure. It also enhances your grasp of interconnections within the model. Deepening Model Structure Understanding Visualize dependency chains to improve documentation clarity. This aids in better model maintenance and understanding. Visualizing Dependency Chains Use the dependency insights to assess the impact of changes. For example, determine which measures break if a column is removed. Assessing Impact of Changes Identify both direct and indirect dependencies between model objects. This helps in understanding how different components are interconnected. Identifying Dependencies Join the INFO.CALCDEPENDENCY output table to itself multiple times. This step reveals complex relationships between model objects. Joining Tables for Dependency Analysis Understanding Dependency Chains 18
19. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents Automate and schedule these notebooks for continuous documentation updates. This ensures your data extraction and documentation processes are streamlined and up-to-date. Automating and Scheduling Notebooks Extract data into pandas or Spark DataFrames for further processing. This step ensures your data is ready for advanced analysis and manipulation. Extracting Data for Processing List datasets, tables, and measures; run DAX queries like INFO.CALCDEPENDENCY directly. This allows for efficient exploration and querying of your data. Exploring and Querying Data Connect to your workspace and dataset programmatically. This enables seamless interaction with your data sources. Connecting to Workspace and Dataset Install SemPy in your notebook environment. This is the first step to begin automating data extraction with SemPy. Installing SemPy Automating Data Extraction with SemPy 19
20. 01 02 04 03 Automating Power BI Documentation with Microsoft Fabric Data Agents Enable agents to query this data efficiently for up-to-date documentation generation. This step is crucial for building an automated, scalable documentation workflow. Efficient Data Querying Maintain a single source of truth for model dependencies and metadata. This guarantees consistency and reliability in documentation processes. Single Source of Truth Centralize documentation data for easy access by Fabric Data Agents. This facilitates streamlined workflows and enhances collaboration. Centralized Documentation Access Save SemPy DataFrames as Parquet files or tables in your Lakehouse storage. This ensures efficient storage and retrieval of extracted data. Saving SemPy DataFrames Saving Extracted Data to Lakehouse 20
21. Automating Power BI Documentation with Microsoft Fabric Data Agents Adding synonyms and metadata helps the agent interpret your data correctly. This step improves the quality and reliability of AI-generated documentation. Adding Synonyms and Metadata Ensuring data is well-structured and free of errors enhances its reliability. Proper structuring minimizes the risk of misinterpretation by AI systems. Error-Free Data Structuring Using clean, consistent naming conventions for tables and columns is crucial. This helps maintain clarity and reduces confusion during data interpretation. Consistent Naming Conventions Improve agent accuracy by removing unnecessary brackets and formatting quirks from your data before saving. These practices ensure the agent can process data more effectively and accurately. Enhancing Agent Accuracy Best Practices for Data Preparation 21
22. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents Save and test the instructions to ensure the agent understands and executes them properly. Testing is crucial to verify the accuracy and effectiveness of the agent's output. Save and Test Instructions Optionally, include instructions in the model’s LSDL for finer control. This provides additional flexibility in tailoring the agent's behavior. Utilize LSDL for Advanced Control Write clear and natural language instructions, such as 'Generate documentation for measures in the CalcDependency table, describing each calculation.' These instructions guide the agent in producing meaningful and specific documentation. Craft Natural Language Instructions Enable the 'Enable for query' option on relevant tables. This step allows the agent to interact with the tables effectively during queries. Activate Query Options for Relevant Tables Add your documentation tables and metadata to the agent configuration. This ensures the agent has access to the necessary data for generating accurate outputs. Integrate Documentation Tables and Metadata Configuring AI Instructions for Your Agent 22
23. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents Address permissions or data access issues if results are unexpected. Resolving these issues is essential for maintaining agent functionality and data accuracy. Troubleshooting Unexpected Results Refresh your Power BI model after any schema or instruction changes to sync updates. This action ensures the agent reflects the latest data structure and instructions. Syncing Updates with Power BI Check that returned documentation is clear, relevant, and complete. This step is crucial for maintaining the reliability and usability of the agent's outputs. Evaluating Documentation Quality Confirm the agent accesses the correct tables and interprets instructions accurately. This ensures the agent retrieves relevant data and performs tasks as expected. Verifying Data Access and Interpretation Run test queries in the Fabric workspace chat interface to ensure the agent operates as intended. This step helps verify the agent's ability to interpret instructions and access the correct tables. Testing Queries in Fabric Workspace Testing and Validating Your Data Agent 23
24. Automating Power BI Documentation with Microsoft Fabric Data Agents These practices help both the agent and Copilot perform more effectively. They contribute to a more accurate and user- friendly query experience. Enhanced Agent and Copilot Performance Good naming and synonyms reduce ambiguity in AI-generated content. This fosters trust and confidence in the system's outputs. Boosting User Confidence Using synonyms and naming conventions improves the relevance and clarity of generated explanations. This ensures that AI-generated content aligns better with user expectations. Improved Query Interpretation Adding synonyms for tables, columns, and measures captures business terminology variations. This approach helps both the agent and Copilot interpret queries more accurately. Incorporating Synonyms Applying clean, descriptive, and consistent naming conventions across your data model enhances agent understanding. This practice ensures clarity and reduces ambiguity in interpreting queries. Consistent Naming Conventions Using Synonyms and Naming Conventions 24
25. Automating Power BI Documentation with Microsoft Fabric Data Agents Monitoring Agent Health and Performance 25 01 02 03 04 05 Proactive Issue Resolution Proactively addressing issues avoids downtime or inaccurate documentation. This ensures the automated documentation process remains dependable and uninterrupted. Audit Logs for Security Compliance Regularly reviewing audit logs ensures adherence to security protocols. This proactive approach mitigates risks and enhances system trustworthiness. Query Performance and Error Tracking Tracking query performance and error logs helps identify inefficiencies. It allows for timely resolution of issues to maintain system reliability. Continuous Monitoring of Permissions Monitoring permissions and data accessibility continuously is crucial. This practice prevents unauthorized access and ensures data integrity. Diagnostic Tools for Agent Status Maintain smooth operation with diagnostic tools to check agent status and readiness. These tools ensure agents are prepared and functioning optimally for automated documentation.
26. Automating Power BI Documentation with Microsoft Fabric Data Agents 01 02 03 04 This interface simplifies access to complex documentation. It enables users to interact with data models without technical barriers. Empowering Users Use this interactive approach for quick insights. Ad hoc explanations are easily accessible through natural language queries. Interactive Insights The agent processes your request and returns detailed, easy-to-understand documentation. This user-friendly interface democratizes access to complex model documentation. Processing Requests Open the Fabric workspace chat interface. Type natural language requests like “Please generate documentation for the measures in the CalcDependency table.” Initiating Interaction Generating Documentation via Microsoft Fabric UI 26
27. Automating Power BI Documentation with Microsoft Fabric Data Agents Use the provided Python snippet to generate and save documentation. This approach embeds documentation efforts into your data operations effectively. Example Code Snippet Set up schedules to keep documentation current without manual intervention. This reduces the risk of outdated information and saves time. Schedule Regular Updates Incorporate documentation workflows directly into your scripts and tools. This ensures that documentation processes are automated and consistent. Integrate Workflows into Tools Use Python notebooks to send documentation generation requests programmatically. This enables seamless integration of documentation tasks into your data workflows. Leverage Python Notebooks Automating Documentation with Fabric Data Agent SDK 27
28. 01 02 03 04 05 Automating Power BI Documentation with Microsoft Fabric Data Agents Archiving and Collaboration Download files locally or upload to documentation platforms like GitHub. Exporting enables easy archiving, collaboration, and distribution of documentation. Customizing Output Templates Customize output templates with titles, metadata, and headers to match your team’s style. This ensures the exported documentation aligns with organizational branding and requirements. Benefits of CSV Format CSV is suitable for spreadsheet analysis or integration with other tools. This format is ideal for data-driven tasks and compatibility with analytical software. Advantages of Markdown Format Markdown offers readable, formatted documents ideal for sharing and version control. It is a preferred choice for teams focusing on collaborative editing and tracking changes. Export Options in Fabric Workspace After generating documentation, use the Fabric workspace export options to save files as Markdown or CSV. These formats cater to different needs, ensuring flexibility in documentation handling. 28 Exporting Documentation to Markdown and CSV
29. Automating Power BI Documentation with Microsoft Fabric Data Agents Best Practices for Documentation Version Control 29 Accurate documentation aligns with evolving models and processes. This approach keeps information traceable and up-to-date. Ensure Documentation Alignment Work with your team to refine and enhance documentation. Team collaboration ensures alignment with evolving models and needs. Collaborate for Continuous Improvement Track changes over time for rollback capability. This approach provides a clear history of modifications for audit purposes. Track Changes for Auditability Use pull requests and code reviews to uphold consistency. Collaborate with your team to continuously improve documentation. Maintain Quality with Pull Requests Store Markdown files in version control systems such as GitHub. This ensures documentation remains accurate and traceable over time. Treat Documentation Like Code 01 02 03 04 05
30. 01 02 04 03 Automating Power BI Documentation with Microsoft Fabric Data Agents Set up automated refreshes and notifications to deliver the latest updates. This promotes transparency and keeps everyone aligned with the most current data. Automated Updates and Alerts Distribute documentation via Power BI Apps to reach large or external audiences. This approach helps in sharing insights beyond your immediate team. Reaching Broader Audiences Embed Power BI reports and documentation in Teams or SharePoint for easy access. This ensures that critical information is readily available to your team. Centralized Access to Resources Encouraging Team Discussions Post Markdown files in Microsoft Teams channels to encourage discussion. This keeps stakeholders informed and engaged in the documentation process. Sharing Documentation with Your Team 30
31. Organize deployments with pipelines for consistent, repeatable processes. This method supports efficient and reliable operations across workspaces. Streamlined Deployments Manage access with security groups to control permissions at scale. This ensures robust and secure handling of sensitive data. Scalable Security Management Automate documentation generation for every Power BI model in your environment. This approach keeps your records accurate and up-to-date effortlessly. Comprehensive Documentation Enterprise-Wide Automation Expand automation enterprise-wide by using PowerShell scripts or Power BI REST API to loop through multiple workspaces. This ensures seamless integration and scalability across your organization. Scaling Documentation Automation Across Workspaces 31 Automating Power BI Documentation with Microsoft Fabric Data Agents