2026 Update: Build an HR Copilot Agent Companion Guide

This report serves as the April 2026 companion piece to Kelly Edinger's mid-2024 session, "Build an HR Copilot Agent." The Microsoft Copilot Studio platform has undergone significant architectural changes since that recording. As a senior consultant updating a colleague, my goal is to bridge the gap between what you watched and how we execute these builds today.

The original session demonstrated manual topic orchestration and separate application lifecycle management. Today, the platform favours generative orchestration, embedded solution management, and native tool calling. This guide translates Kelly's exact demonstrations into current 2026 best practices.

What's Changed Since This Session

The landscape of Microsoft Copilot Studio has matured significantly over the last two years. The terminology, underlying architectural patterns, and licensing models have evolved to support enterprise-scale generative AI. The following table provides a factual summary of these critical shifts.

How to Build This Today

The original session showcased six distinct demonstrations covering the end-to-end creation of an HR agent. This section details exactly how you would build those same scenarios using the Microsoft Copilot Studio platform as it exists in April 2026.

Scenario 1: Application Lifecycle Management (ALM) and Environments

The session showed Kelly creating a Power Platform solution with Dataverse for application lifecycle management (ALM) to separate dev from production environments. She had to leave the conversational interface and navigate to the external Power Apps maker portal. Today, this external navigation is no longer required.

In April 2026, Copilot Studio embeds the solution explorer directly within the authoring environment. Agents are automatically provisioned within a default solution, but enterprise deployments require custom solutions. These custom solutions act as transport containers for your agents across different environments.

To build this ALM structure today, open your Copilot Studio environment and look to the sidebar navigation. Select the three dots (...) on the left menu, and then select Solutions. This opens the native solution manager without forcing you into a new browser tab.

Quick Win: Do not rely on the default solution for enterprise agents. Immediately create a custom solution to ensure your agent components, environment variables, and connection references are properly grouped for future export.

Once the solution explorer is open, select New solution from the top menu bar. The system prompts you to define your solution requirements, including a display name, publisher, and version number. After the solution is created, it should open automatically in the explorer list.

You must then instruct the environment to use this specific container by default for all new components. Select Set preferred solution from the top menu. Choose your newly created custom solution from the dropdown list.

All newly created agents will now automatically reside in this managed container. This simple setup future-proofs your HR agent for multi-environment deployments. When the HR agent is ready to move from testing to production, you no longer need to manually export and import ZIP files unless you choose to. Simply open your custom solution and select Pipelines from the side list to configure continuous integration and continuous delivery (CI/CD) pipelines directly from this menu.

These native pipelines allow you to execute single-click deployments to target environments. For advanced developers, Microsoft also released a Visual Studio Code extension to manage these agent solutions directly within the IDE.

Scenario 2: Knowledge Attachment (SharePoint vs Dataverse)

The session showed Kelly comparing knowledge attachment: uploading HR policy files directly to Dataverse vs. linking to a SharePoint document library. She discussed pros and cons like licensing and auto-updates. Here is how you would evaluate and build that same thing in April 2026.

Today, there are two distinct methodologies for unstructured knowledge attachment.

Option 1 (Dataverse File Upload) copies your HR policy files from SharePoint directly into Microsoft Dataverse. Once copied, the system processes these files into semantic indexes and creates vector embeddings. This enables high-quality semantic search across your documents, including reading text within images such as scanned PDF files. However, this option consumes your Dataverse storage capacity, which carries cost implications. Furthermore, Option 1 operates on a batch cycle for updates. When an HR manager updates a policy in SharePoint, that change is not immediately reflected in the agent. The system synchronises changes from the source files every four to six hours. If real-time accuracy is critical for your HR policies, this delay poses a risk.

Option 2 (SharePoint Connector) utilises the native SharePoint Connector to avoid these delays. This approach leaves the files resting natively in SharePoint and does not consume any Dataverse storage. When the agent receives a query, it directly leverages the SharePoint search infrastructure to find the answer.

Warning: Option 2 (SharePoint Connector) does not currently support searching within SharePoint lists. It is strictly limited to document libraries, wikis, and site content.

Option 2 guarantees real-time content freshness, reflecting the latest available document updates immediately. It also supports advanced query filters based on metadata, such as filtering by author, modified date, or title. For most modern HR agents, we recommend Option 2 due to its real-time synchronisation and lack of storage overhead.

To implement Option 2, you must ensure your users have the correct permissions. The system performs a live authorisation check on the user's Entra ID connection information (such as Sites.Read.All and Files.Read.All) at the source system. Files with sensitivity settings of "Confidential" or "Highly Confidential" are automatically blocked from agent answers.

Scenario 3: Designing Conversational Topics

The session showed Kelly designing conversational topics in Copilot Studio. She started with basic Q and A on HR policies and transitioned into interactive time-off requests where the agent explicitly asked for details. Here is how you would build that same conversational flow in April 2026.

The entire paradigm of conversation design has shifted from classic orchestration to generative orchestration. Traditional topic-driven designs required makers to anticipate every user intent and build rigid, manual branching logic. The agent previously relied on handcrafted "trigger phrases" to route users correctly.

Today, generative orchestration introduces a large language model (LLM) acting as a dynamic, central planning layer. This planner interprets the user's intent, breaks down complex requests, and automatically selects the correct topics and tools. You no longer need to build overlapping logic or exhaustive trigger phrase lists.

To build an interactive time-off request today, you create a topic but rely on the LLM for slot filling. The system no longer requires manual question nodes to explicitly ask the user for missing data piece by piece. Instead, you simply define the required inputs for your time-off action (such as Start Date, End Date, Reason).

When the user types, "I need to take next Friday off," the generative orchestrator recognises the underlying intent. The planner automatically maps "next Friday" to the required Start Date and End Date slots using natural language understanding. It then recognises that the "Reason" slot is still empty and dynamically generates a natural, conversational question to ask the user for the missing reason -- without the developer writing a single line of script or building a new question node.

Quick Win: Ensure Generative Orchestration is explicitly enabled in your agent's settings. Provide clear, comprehensive descriptions for every topic and tool, as the LLM relies entirely on these textual descriptions to determine routing.

Furthermore, generative orchestration handles multi-intent utterances that would break classic agents. A user can say, "Check the HR policy on bereavement and then submit a request for tomorrow." The LLM processes both requests and executes them sequentially.

If you prefer to work with code, Copilot Studio now features a native YAML code editor for topic design. This allows developers to view and edit topics in a readable markup language and copy complex YAML configurations between different agents.

Scenario 4: Customising Topics and Preventing Hallucinations

The session showed Kelly customising out-of-the-box topics, adding new custom topics, and refining instructions to control agent behaviour. Her goal was to prevent hallucinations and keep the agent focused on HR. Here is how you would secure and instruct that agent in April 2026.

Today, securing the agent requires robust, multi-layered generative AI guardrails. This configuration begins globally within the Generative AI tab of the agent settings. Here, you set the absolute boundaries for knowledge retrieval, define the permitted website URLs and uploaded documents, and set the desired moderation level for the entire agent.

To refine specific topic flows, you must utilise the Generative answers node within the authoring canvas. Adding custom instructions to this specific node yields the highest return on investment for preventing hallucinations.

To configure this, open your topic and select the three dots (...) on the Generative answers node. Open the Properties pane, and locate the Custom instructions field. Provide precise, assertive instructions here.

For an HR agent, your instructions must explicitly state the agent's persona and limitations. You must instruct the LLM to maintain a professional, supportive tone, strictly avoid fabricating facts, and proactively ask clarifying questions when information is missing rather than guessing.

Warning: User prompts undergo strict safety filtering, but maker-configured system instructions do not. You carry total responsibility for ensuring your custom instructions do not introduce security vulnerabilities or encourage ungrounded, hallucinatory responses.

To further prevent context errors, you should employ dynamic prompts. Copilot Studio allows makers to use Power Fx formulas directly within the custom instructions field. This capability enables the instructions to adapt automatically based on variables established earlier in the conversation. For example, if the agent detects the user is a manager, the Power Fx formula can dynamically inject instructions to reference leadership guidelines.

You can also implement Data Loss Prevention (DLP) enforcement across your agents using PowerShell commands to block unauthorised updates.

Scenario 5: Implementing Tools for SharePoint Lists

The session showed Kelly implementing 'actions' to post time-off request data from conversations directly into a SharePoint list. This enabled subsequent Power Automate approval flows. Here is how you would build that same connection in April 2026.

In current terminology, 'actions' have been formally rebranded and expanded as 'Tools'. Tools serve as the primary mechanism for an agent to execute tasks and interact with enterprise systems. Natively calling a tool is significantly faster than designing a custom Power Automate flow for simple data entry.

To build this SharePoint connection today, open your agent and navigate to the Tools page from the left-hand navigation pane. Select Add a tool at the top of the screen. Choose Connector from the list of available tool types.

The system presents a search box; enter "SharePoint" and select the standard connector. The interface reveals all available operations for SharePoint. Locate and select the Create Item tool. This tool allows the agent to push new rows into a designated list directly from the chat.

Once added, you must configure the tool parameters. You will provide the specific SharePoint Site Address and the List Name. The interface dynamically exposes the columns of that list (such as Title, Start Date, End Date, Reason).

Because you defined these slots in your instructions, the LLM automatically maps the extracted conversational entities directly into the SharePoint Tool inputs. You simply save the tool and publish the agent. When a user requests time off, the LLM planner gathers the required data through natural conversation, securely invokes the SharePoint Create Item tool using the user's Entra ID context, and confirms the successful database entry back to the user in natural language.

If you require advanced security for tools, Copilot Studio now supports Model Context Protocol (MCP) servers. MCP servers enable secure, enterprise-grade tooling driven by natural language, suitable for complex HR logic apps.

Scenario 6: Full Agent Setup and Jumpstart Learnings

The final segment showed Kelly setting up the full agent with descriptions and sharing Jumpstart program learnings on pitfalls like dynamic content issues. Here is how you would execute full setup and avoid those pitfalls in April 2026.

Today, complete agent assembly is accelerated. Microsoft has released an Employee Self-Service Agent template that provides a head start specifically for HR scenarios. This template includes prebuilt connectors and starter workflows for leave management. Verify current integration availability with HRIS systems like Workday and SAP SuccessFactors against the current Copilot Studio template gallery, as available integrations change with each wave.

If you choose to build from scratch, you start on the Copilot Studio Home page. The interface supports natural language creation. You simply type, "Create an HR agent to answer policy questions and manage time-off requests."

The AI automatically provisions the base framework, generates a suggested name, description, and initial instructions based on your prompt, and suggests relevant triggers and knowledge sources.

Quick Win: Always manually refine the AI-generated instructions. Clear, concise, and boundary-defining system prompts remain the most critical component of a reliable, compliant HR agent.

The Jumpstart program learnings regarding dynamic content remain relevant. Historically, developers struggled to parse unstructured or dynamic JSON payloads returned by custom APIs. Makers frequently resorted to brittle string-scraping techniques to extract data for the conversational interface.

Today, the recommended architecture avoids string scraping entirely. Copilot Studio now supports native schema-based JSON parsing. When your agent retrieves complex backend data, the parsed output instantly becomes typed dynamic content. You can simply reference structured tags directly within your conversational response configuration. This ensures the agent correctly formats complex data, such as a summary of available leave balances, without hallucinating formatting errors.

Licensing Quick Reference

The licensing model for Microsoft Copilot Studio changed in September 2025 when the billing currency shifted from "messages" to Copilot Credits. The table below outlines the primary licensing scenarios for deploying an HR agent.

Key billing facts:

• As of September 1, 2025, Copilot Studio uses Copilot Credits as the billing currency, replacing the old per-message model. The quantity per prepaid pack did not change.
• When an internal HR agent is deployed to Teams, SharePoint, and Microsoft 365 surfaces, and used by M365 Copilot-licensed users, those interactions are included at no additional charge (subject to fair use limits).
• If you deploy the HR agent to an external website or allow unauthenticated or unlicensed users to access it, those interactions consume Copilot Credits.
• Agent actions (such as invoking a SharePoint tool during a conversation) are billed at 5 Copilot Credits per action for unlicensed users.
• To forecast your Copilot Credit consumption before launch, use the Microsoft Copilot Studio agent usage estimator.
• For large-scale deployments, prepaid capacity packs offer a predictable billing model. Contact your Microsoft licensing representative or check the current Copilot Studio Licensing Guide for current pack sizes and pricing, as these figures change.

Note on pricing figures: Specific dollar amounts for Copilot Credit prepaid pack tiers were not included in this guide because published pricing can change between Microsoft licence waves. Always verify current pricing in the Microsoft Copilot Studio Licensing Guide or the Microsoft admin centre before committing to a capacity purchase.

Evidence Notes

Generative orchestration in Copilot Studio is a confirmed, available feature (not preview for standalone Copilot Studio environments). The Microsoft Copilot Studio What's New page and the 2026 Wave 1 release plan confirm its availability.

The billing currency change from messages to Copilot Credits (September 1, 2025) is confirmed by the official Copilot Studio Licensing documentation.

Agent action billing rate (5 Copilot Credits per action) is confirmed by the official Copilot Studio billing rates table.

The Employee Self-Service Agent template is referenced as a Microsoft release. Verify current availability and HRIS integration support in the Copilot Studio template gallery, as available templates change with product waves.

Specific pricing figures for Copilot Credit prepaid capacity tiers (previously shown as "$2,850.33" in earlier versions of this guide with a footnote number fused into the price) have been removed. These figures were garbled in the original and not verifiable against primary Microsoft sources at the time of this repair.