AI Assistants
BI Pixie records how people actually use your Power BI reports, but that evidence is only useful if the people and tools making decisions can reach it. Reading it through the dashboard means opening a report and knowing which page to look at, and questions that fall outside the prepared visuals ("which reports has this team stopped opening since the redesign?") require someone to build a new view first. AI assistants remove that step: they answer engagement questions in plain language, and they can bring that context into work already under way, such as deciding what to retire, where to invest, or which audience to survey.
On the AI Assistants page, you connect your AI tools to your BI Pixie semantic model, so that people and AI agents alike can explore report engagement, adoption, and user satisfaction, grounded in the same data as your BI Pixie Dashboard. There are two options:
- Fabric data agent: a native, zero-install data agent that everyone in your workspace can chat with. People ask it questions in plain language inside Microsoft Fabric and Microsoft Copilot. It suits business users best.
- BI Pixie MCP Server: connect your own AI tools (Claude, GitHub Copilot, Codex, Microsoft Foundry) for exact, repeatable, programmatic access. It suits developers, analysts, and automation best.
AI Assistants is in the left sidebar of your BI Pixie item. You can use either option, or both. The sections below cover each option in turn.
To use BI Pixie in claude.ai and your other Claude apps with nothing to install, the Use BI Pixie in Claude card on the AI Assistants page gives the connector address for your account. After your Claude has connected, the card says since when, and the address is under Add BI Pixie to Claude again. See Use BI Pixie in Claude.
To use BI Pixie in Copilot, in Teams or the Copilot app, you add BI Pixie from Microsoft's Agent Store. See Use BI Pixie in Copilot.
Compare the two
| Fabric data agent | BI Pixie MCP Server | |
|---|---|---|
| Where it runs | Inside Microsoft Fabric | On your own machine, or hosted in your Azure environment |
| Ask from | Microsoft Fabric, Microsoft Copilot, and Copilot Studio, plus Claude, GitHub Copilot, and Codex via the data agent's preview MCP server | Claude, GitHub Copilot, Codex |
| Best for | Business users, developers, and AI coding tools | Developers and AI coding tools |
- You can deploy the MCP server remotely with the BI Pixie managed app, and run it as a secure, always-on endpoint for production AI agents and teams. The Hosted option section below describes it.
- Use a data agent as an MCP server: how to expose your Fabric data agent as its own MCP endpoint.
- Use a data agent in Copilot: how to chat with your data agent directly from Copilot.
Fabric data agent
A Microsoft Fabric data agent answers questions in plain language, grounded on your BI Pixie semantic model. Anyone in your Fabric workspace can ask about report usage, adoption, and satisfaction in plain language, right inside Microsoft Fabric and Microsoft Copilot, with nothing to install. It is the most direct way to put BI Pixie's engagement data in front of non-technical colleagues.
Add a data agent
- From your BI Pixie item, open the AI Assistants page from the left sidebar.
- On the Add an AI Data Agent card, select Add data agent.
BI Pixie creates the data agent in your workspace, in the same BI Pixie folder as your dashboard, and grounds it on your semantic model automatically, with no configuration. Adding the data agent requires:
- Permission to create a data agent in the workspace (a Contributor or Member role with the create-item right). Without that permission, BI Pixie tells you on the card to ask a workspace admin.
- Your tenant's Copilot and Azure OpenAI setting enabled by a Fabric administrator.
- A workspace on a paid Fabric capacity, such as F2 or higher, or P1. Fabric does not run data agents on a trial capacity.
- An installed BI Pixie Dashboard, the semantic model the data agent grounds on.
Open and publish it
Once BI Pixie has created the data agent, the card shows Open data agent, and the same action appears in the BI Pixie ribbon. A new data agent answers questions only after you open it in Fabric and publish it in the data agent editor. After that, you and your colleagues can chat with it from Fabric or Microsoft Copilot.
Use it as your AI
The data agent has a second role: it can serve as your AI in AI Readiness, the AI that writes the descriptions, AI Instructions, folder names, and benchmark question wordings BI Pixie proposes for your semantic models. Choosing it means there is no API key to paste and no separate account to manage, because the data agent runs on your own Fabric capacity.
Once the data agent exists, its card offers Use for optimizations. Selecting it opens the same Set up AI dialog the AI Readiness page uses, so that you can connect the data agent as your AI without leaving this page. BI Pixie queries the data agent with your identity, which needs your permission to query Fabric data agents, a permission you can grant yourself with no tenant administrator required. If you have not granted it, the data agent is still saved as your AI, and the AI Readiness page offers Grant access the first time the permission is needed.
Use for optimizations appears only for admins, because your AI is a shared account setting rather than a personal one. To use your own Microsoft Foundry, OpenAI, or Anthropic account instead, you can choose that service in the Set up AI dialog. See AI Setup.
What you can ask
You can ask in plain language, for example:
- "Which reports are used the most? Which are unused and could be retired?"
- "How has adoption trended over the last 90 days?"
- "What's our CSAT and NPS, and which reports drag satisfaction down?"
- "Which pages get the most clicks and time spent?"
Good to know
- Suited to exploring. The data agent turns your question into a query on the fly. Important numbers are worth checking against your BI Pixie Dashboard. The BI Pixie MCP Server, described below, gives exact, repeatable, and programmatic answers.
- Same data as your dashboard. It grounds on the same BI Pixie semantic model, so its answers come from the same validated measures.
- Treat answers like workspace data. The data agent doesn't apply row-level security or strip personal data, so anyone who can chat with it can see whatever the semantic model exposes. Share it the way you'd share the underlying report.
BI Pixie MCP Server
The BI Pixie MCP Server lets AI assistants (such as Claude, GitHub Copilot, Codex, Microsoft Fabric data agents, and Microsoft Foundry) answer questions about how your Power BI reports are actually used, in plain language. It connects to the BI Pixie semantic model in your Fabric workspace and returns numbers that match your BI Pixie Dashboard exactly. Its usage tools are read-only, so they never change, delete, or refresh your data. Since bipixie-mcp 0.6.0 the BI Pixie MCP Server also carries tools that work in BI Pixie as you, through the BI Pixie API. They come with the local install, and with a hosted deployment whose deployer turned them on. Each of those shows you what it would change and waits for your answer first.
BI Pixie runs natively inside Microsoft Fabric as a BI Pixie item. The MCP Server is a separate tool that you register with your AI assistant. It reaches your engagement data by querying the BI Pixie semantic model that lives in the same Fabric workspace where you installed the dashboard, using your own Fabric and Power BI identity.
About the Model Context Protocol (MCP)
The Model Context Protocol (MCP) is an open standard for connecting AI assistants to tools and data. The BI Pixie MCP Server is one such tool: once you register it with your AI assistant, you can ask usage questions in chat and the assistant pulls the answers straight from your BI Pixie data. You write no DAX and export nothing.
Two ways to run it
You can run the BI Pixie MCP Server in two ways. The local install is the place to start, and the remote option is how you run the server in production for AI agents and teams.
- Local (start here). You install the server on your own machine and sign in with your Fabric and Power BI identity, with no service principal and no admin setup beyond one tenant setting. The steps below cover the local install.
- Remote (run it in production). Your IT or Azure team hosts the server as a secure, always-on endpoint, so that AI agents in production workflows can query usage data on their own, and so that Microsoft Fabric data agents, Microsoft Foundry, and your team can reach it with no local install. See the Hosted option section below.
Ground your data agent in real human usage
When a data agent turns a question into DAX against a mission-critical semantic model, it has to guess which column and which value the user means. The BI Pixie MCP Server can answer that question from real usage before the data agent writes its query. With data_audit and group_by="column", BI Pixie ranks columns by how often people actually select and filter them. With data_audit and view="column_values", BI Pixie returns a column's most-selected values. The data agent then picks the column and the values real users rely on instead of guessing. Grounding in real usage matters most in a hosted, always-on deployment where data agents run on their own. Both answers count the data-selection events BI Pixie captures (slicer, filter, cross-filter, and drill selections) rather than raw clicks on individual visuals, and every result carries a freshness footer.
What you can ask
Once the server is connected, you can ask your assistant questions like:
- "Which of my reports are used the most? Which are unused and could be retired?"
- "How many monthly and daily active users do I have, and what's my stickiness (DAU/MAU)?"
- "Which pages in the Executive Dashboard get the most clicks and time spent?"
- "What's my CSAT and NPS, and which reports are dragging satisfaction down?"
- "How has adoption trended over the last 90 days?"
- "How much time savings and financial benefit have users reported in surveys?"
- "Users keep asking about 'region'. Which audited column do people actually filter on most, so my data agent groups by the right one?" (BI Pixie resolves the name, then ranks columns by real selection activity.)
- "Before I write the DAX, what are the most-selected values of the Country column, the ones people deliberately click, so I can default my filter to them?" (BI Pixie ranks the column's values by what people actually select.)
Tip: when you ask your assistant to "suggest questions", BI Pixie returns a built-in menu of example questions grouped by topic, so you can explore what's available.
Before you start
- An active BI Pixie license and a deployed BI Pixie semantic model. This is the same semantic model behind your BI Pixie Dashboard, installed in your Fabric workspace when you set up the dashboard.
- Python 3.12 or later on your machine.
- The Azure CLI, for the simplest sign-in (
az login). - One Power BI admin setting: "Dataset Execute Queries REST API" enabled (Power BI Admin portal > Tenant settings > Integration settings). This is the single tenant prerequisite for the local install. If it's off, queries fail with a permission error. No service principal and no "Allow service principals to use Power BI APIs" toggle are needed for the local path.
- Normal Power BI workspace access for your signed-in account (Viewer, Member, or whatever the semantic model needs), granted the usual way in Manage access on the Fabric workspace. This is standard per-user access, not a service-principal enrollment.
- Three values that identify your data: your Microsoft Entra tenant, the Fabric workspace that holds the BI Pixie semantic model, and the semantic model name (
BI Pixie Dashboard). You can use the workspace's display name instead of its GUID.
Step 1: Install
Install in one click from AI Assistants
If you have set up the BI Pixie Dashboard, you can connect from inside Microsoft Fabric with Install in VS Code on the Use BI Pixie MCP Server card, on the AI Assistants page in the left sidebar of your BI Pixie item. That link is already filled in with your Fabric workspace, so you can skip the manual configuration in Step 2. You still need Python (Step 1) and a one-time az login. To configure it by hand, follow the steps below.
You install the server from PyPI, with no download or clone required.
# Zero-install runner (recommended), always fetches the latest version:
uvx bipixie-mcp
# Or install into your current Python environment:
pip install bipixie-mcp First run walks you through it
When you run uvx bipixie-mcp in a terminal with no settings yet, BI Pixie starts a short guided setup. It signs you in (reusing az login when available), lists the workspaces you can access, flags your BI Pixie semantic models, lets you pick one from a numbered menu, runs a quick test query to confirm it works, and saves a .env you can reuse. You can run the guided setup again any time with uvx bipixie-mcp setup. To configure it by hand, skip ahead to Step 2. (When your AI assistant launches the server, the server never prompts. It reads the settings described in Step 2.)
In VS Code, you can register the server in one click and then add your settings in Step 2:
Step 2: Configure
The server reads its settings from environment variables (or from your AI assistant's MCP config). Point the workspace and semantic model at the Fabric workspace where you installed the dashboard and at the BI Pixie Dashboard semantic model inside it. The simplest setup uses your existing Azure CLI sign-in, with no app registration and no secret. Create a .env file with these values. The last one, the BI Pixie API's application ID, turns on the tools that work with BI Pixie itself, described in Manage BI Pixie from your assistant. Leave it out for the read-only usage tools alone.
BIPIXIE_MCP_TENANT_ID=<your-entra-tenant-guid>
BIPIXIE_MCP_WORKSPACE_NAME=<Fabric workspace that holds the BI Pixie semantic model>
BIPIXIE_MCP_DATASET_NAME=BI Pixie Dashboard
BIPIXIE_MCP_AUTH_MODE=azure_cli
BIPIXIE_MCP_API_APP_ID=114a7e20-3796-4ddd-ab99-568a78b7759b Then sign in once. The server reuses that session for every query:
az login Tip: find your tenant GUID with az account show --query tenantId -o tsv. To use GUIDs, set BIPIXIE_MCP_WORKSPACE_ID instead of the workspace name.
You do not need to look up IDs. The settings above use display names, so you do not have to copy GUIDs from the Fabric or Power BI portal. After the server connects, you can ask your assistant to list your Fabric workspaces and BI Pixie semantic models and pick one, then switch between them at any time.
Sign in without the Azure CLI
Set BIPIXIE_MCP_AUTH_MODE=device_code instead. This sign-in mode needs a public-client Entra app registered in your own tenant, with its client ID in BIPIXIE_MCP_CLIENT_ID. The first time you use the server, it prints a short sign-in code to paste into your browser. Either way, the server runs as you. There is no separate BI Pixie sign-in step: Microsoft Fabric and Power BI provide your identity automatically, and the server can only reach data you already have access to in Power BI. Neither local mode needs a service principal. The tools that act on BI Pixie, described in Manage BI Pixie from your assistant, sign in with the Azure CLI.
Step 3: Register with your assistant
Claude Code, Claude Desktop, or Cursor
In Claude Code and Cursor, add the server to your project's .mcp.json. Claude Desktop keeps its servers in its own configuration file instead: in Claude Desktop, open Settings, then Developer, and select Edit Config. The entry is the same in each:
{
"mcpServers": {
"bipixie": {
"command": "uvx",
"args": ["bipixie-mcp"],
"env": {
"BIPIXIE_MCP_TENANT_ID": "<your-entra-tenant-guid>",
"BIPIXIE_MCP_WORKSPACE_NAME": "<your-fabric-workspace-name>",
"BIPIXIE_MCP_DATASET_NAME": "BI Pixie Dashboard",
"BIPIXIE_MCP_AUTH_MODE": "azure_cli",
"BIPIXIE_MCP_API_APP_ID": "114a7e20-3796-4ddd-ab99-568a78b7759b"
}
}
}
} VS Code
Select Install in VS Code above, then add the same env block to .vscode/mcp.json so the server knows which Fabric workspace and semantic model to query.
Restart your assistant. Ask "describe the BI Pixie semantic model" to confirm the connection. You should see your tables, measures, and the data's freshness date.
Then ask "What is my BI Pixie plan?" to confirm that the BI Pixie tools work as you. You should see your account, your plan, and what is left of your allowances today and this month. If the first answer works and the second does not, check BIPIXIE_MCP_API_APP_ID and that you signed in with az login.
Switch workspaces while you chat
If you manage BI Pixie items across several Fabric workspaces, you don't need to reconfigure to move between them. On the local install, you can ask:
- "List my Power BI workspaces."
- "List the semantic models in <workspace>." (BI Pixie semantic models are flagged for you.)
- "Use the BI Pixie semantic model in <workspace>."
The switch lasts for the session and resets when you restart. You only ever see Fabric workspaces and semantic models your own Power BI account can already access. (This applies to the local install; a hosted deployment stays pinned to one semantic model.)
Manage BI Pixie from your assistant
Beside the usage tools, the server can carry a second set of tools that work with BI Pixie itself, through the BI Pixie API. They act as you, with your own permissions and your own plan, and within the same limits your BI Pixie item applies. Every plan, Free included, can use them.
You can ask your assistant to read what BI Pixie knows:
- "What plan am I on, and how many assessments do I have left today?"
- "Which semantic models has BI Pixie assessed, and what is each one's AI Readiness score?"
- "Show me the last benchmark run on the Sales semantic model, and which AI answered it."
- "Which of my reports have Pixies?"
And to act for you:
- Assess a semantic model, several semantic models, or a data agent.
- Show what BI Pixie would change on a semantic model or a data agent, write it, and undo it. You can give your own wording for a description or for AI instructions in place of what BI Pixie proposed.
- Add Pixies to a report, or remove them.
- Write benchmark questions, save them, and run the benchmark.
- Create and run an A/B test, schedule assessments, and stop a run.
You are asked first
Anything that spends an allowance, claims a tracked item, changes an item, or deletes a result starts as a preview, and nothing is done yet. The preview names each item as your BI Pixie item shows it, for example Assess the AI Readiness of the semantic model "Sales" in the workspace "Finance", says what it would use from your allowances in the numbers your BI Pixie item shows, and ends with the item IDs so two items with the same name can be told apart. When your AI assistant can ask you a question on the server's behalf, it asks you there. Otherwise the assistant shows you the preview and acts only after you agree. An agreement is good once, for that one action, for ten minutes. Showing what BI Pixie would change spends nothing and changes nothing, so it runs without asking.
How you are asked is your choice
Set BIPIXIE_MCP_CONFIRM in the same env block to choose how BI Pixie asks you:
ask, the default: your AI assistant asks you when it can, as described above.chat: your AI assistant is never asked to put a question to you. The preview always comes back into the conversation, and BI Pixie acts once you agree there.off: nobody is asked. Each action happens at once, and its answer says what was done and what it used.
The choice is a setting in the local install only. No tool accepts it, so your AI assistant cannot choose to ask you less. A hosted deployment always uses chat, and cannot be set to off.
Turn them on
Sign in with the Azure CLI (az login) and add one setting, the BI Pixie API's application ID, to the env block from Step 2. The examples above already include it:
BIPIXIE_MCP_AUTH_MODE=azure_cli
BIPIXIE_MCP_API_APP_ID=114a7e20-3796-4ddd-ab99-568a78b7759b One az login covers both sets of tools: nothing needs to be registered, and no administrator is asked to consent. Without the application ID, the server carries the read-only usage tools alone. To keep the BI Pixie tools that read but none that act, also set BIPIXIE_MCP_WRITE_TOOLS=false.
A hosted deployment carries these tools only when its deployer turns them on. It can offer the tools that read alone, or those and the tools that act. Either way they act for the person who signed in to it, never as the deployment itself. The first time a person uses one, they are asked once to allow it, for themselves only. After that, every action is agreed by that person in the conversation, each time, and a hosted deployment cannot be set to ask less. The settings its deployer needs are in the bipixie-mcp reference on PyPI.
Privacy & security
- Usage tools are read-only. They can only run read queries. They can never change, delete, or refresh your data, and write commands are rejected before they reach Power BI.
- BI Pixie tools ask first. A tool that acts on BI Pixie shows you a preview and acts only after you agree, once, for that one action. In the local install only you can change that, with
BIPIXIE_MCP_CONFIRM. A hosted deployment always asks in the conversation. - Runs as you. It uses your own Microsoft Fabric and Power BI identity, so it can only reach data you're already permitted to see. There is no separate BI Pixie sign-in.
- Personal data follows your dashboard. BI Pixie collects viewer user names only when Capture user identity is on in Tracking Setup, and keeps each one as the email address, a masked ID or nothing, according to the choice on the Account page. Otherwise BI Pixie tells viewers apart by a masked IP address. When your BI Pixie semantic model holds user names or IP addresses, the local server returns them as your BI Pixie Dashboard shows them. Set
BIPIXIE_MCP_PII_COLUMNS_ALLOWED=falseto strip them from every result. A hosted deployment strips them by default. - Numbers you can trust. Answers come from the same validated measures as your BI Pixie Dashboard, because both read the same semantic model in your Fabric workspace, so they match. Every answer is stamped with how fresh the data is (the most recent refresh date).
Hosted option: deploy the MCP server remotely (Fabric data agents & Microsoft Foundry)
The same server can run hosted in your Azure environment as a secure, always-on endpoint, so AI agents in production workflows can query usage data on their own, and so Microsoft Fabric data agents, Microsoft Foundry, and your team can reach it without anyone installing anything locally. Unlike the local install, the hosted path runs under a non-user identity, so it uses a service principal or an Azure managed identity instead of your personal sign-in:
- The tenant setting "Allow service principals to use Power BI APIs" (Tenant settings > Developer settings) must be on. This gates only the hosted modes, never the local user-identity install.
- The service principal or managed identity is added as a Member (or Admin) of the Fabric workspace, which gives it Read and Build on the BI Pixie semantic model.
- Managed identity is preferred when hosted: it needs no app registration and no secret. The Azure-assigned identity resolves automatically, and needs the same workspace Member or Admin role as the service principal.
- Service principal is the alternative: a confidential-client app registration with
Dataset.Read.Allas an application permission, admin consent, and a client secret stored as a Key Vault reference.
For the step-by-step deployment, see Deploy the BI Pixie MCP Server, which walks your Azure team through standing up the remote endpoint as part of the BI Pixie managed app. The hosted option is an enterprise setup, and you can contact support to plan it. When its deployer turns them on, a hosted deployment also carries the tools that work with BI Pixie, acting for each person who signs in to it. Manage BI Pixie from your assistant says what each person is asked.
Troubleshooting
- Queries fail with a permission error (403): the "Dataset Execute Queries REST API" tenant setting is off, or your account isn't a Member of the Fabric workspace. Ask your Power BI admin to enable the setting and confirm your workspace access in Manage access.
- An error that begins "No semantic model named": check the Fabric workspace name and confirm
BIPIXIE_MCP_DATASET_NAMEmatches your semantic model's name. BI Pixie names itBI Pixie Dashboardwhen it installs the dashboard. If you renamed it, set the name you gave it. - Wrong data or empty results: confirm
BIPIXIE_MCP_TENANT_IDmatches the tenant where your Fabric workspace lives.
Related Resources
- BI Pixie Semantic Model, the tables and measures the assistant queries, plus the OneLake and Power BI paths to the data.
- AI Readiness, where BI Pixie assesses whether your semantic models are ready for Copilot and data agents, and you fix the gaps it finds.
- Benchmarks, ask an AI real business questions about your semantic model and score its answers against your own data, testing a Fabric data agent, Copilot, or a generic AI agent.
- BI Pixie Metrics, what each measure means.
- BI Pixie Dashboard, a walkthrough of every dashboard page.
- Set up BI Pixie Dashboard, install the semantic model that the MCP Server connects to.
- Data Residency (OneLake), keep your engagement telemetry in your own OneLake.
- bipixie-mcp on PyPI, the full technical reference: every setting, all sign-in modes, the complete tool catalog, and hosted deployment.
- Contact Support, get help with setup.