powerbi-modeling Skill (awesome-copilot) for Claude: What It Does and How to Install It
Makes Claude inspect and improve a Power BI semantic model: star schema, relationships, DAX measures, security roles.
- Installs
- 10K
- on skills.sh
- Stars
- 40.7K
- on GitHub
- Files
- 5
- besides SKILL.md
- Updated
- Oct 2026
- last repo commit
This skill turns Claude into a Power BI modeling assistant. It first connects to your open model and reads its tables, relationships and measures. Then it checks the model against a best-practice list and helps with star schemas, DAX, relationships, performance and row-level security.
It needs the Power BI Modeling MCP server, which does the actual reading and editing. That server can change your model, so back it up first. The skill bundles no scripts, only SKILL.md and 5 reference files.
Good for
- Power BI developers cleaning up or documenting a semantic model
- Analysts writing DAX measures and checking relationships
- Teams setting up row-level security
Install powerbi-modeling
Claude Code
- 1
Copy this prompt.
Install the powerbi-modeling skill in Claude Code. Run this command: npx skills add https://github.com/github/awesome-copilot --skill powerbi-modeling Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Requires Node.js.
The skill only gives Claude modeling instructions. It needs Microsoft's Power BI Authoring MCP server (local) to do anything, so every route below also adds that server. The server runs through npx, so it needs Node.js. The README says you must accept its EULA before the tools work, either by letting your agent call accept_eula after you authorize it, or with --accepteula or PBI_MODELING_MCP_ACCEPT_EULA=true for unattended use. Add --readonly to the server args to block writes. For Fabric workspaces, the README recommends the hosted Power BI Authoring MCP server instead.
Built from the package data we hold. Not yet tested by us.
What's in the skill
When it loads
Power BI semantic modeling assistant for building optimized data models. Use when working with Power BI semantic models, creating measures, designing star schemas, configuring relationships, implementing RLS, or optimizing model performance. Triggers on queries about DAX calculations, table relationships, dimension/fact table design, naming conventions, model documentation, cardinality, cross-filter direction, calculation groups, and data model best practices. Always connects to the active model first using power-bi-modeling MCP tools to understand the data structure before providing guidance.
Sections in SKILL.md
- When to Use This Skill
- Prerequisites
- Required Tools
- Optional Dependencies
- Workflow
- 1. Connect and Analyze First
- 2. Evaluate Model Health
- 3. Provide Targeted Guidance
- Quick Reference: Model Quality Checklist
- MCP Tools Reference
- Common Tasks
- Add Measure with Description
- Update Column Description
- Create Relationship
- When to Use Microsoft Learn MCP
Key facts
- Author
- github/awesome-copilot (repo owner: GitHub; community-contributed); the SKILL.md names no author
- License
- MIT
- Installs on skills.sh
- 9,969 (Nov 2, 2026)
- Last repo commit
- Oct 30, 2026
- Audits on skills.sh
- Gen Agent Trust Hub: pass, Socket: pass, Snyk: pass
What it includes
Lists connections, connects, then reads model, tables, relationships, measures
Uses the Power BI Modeling MCP tools before giving any advice.
Network ·
1. Connect and analyze firstChecks star schema, relationships, naming, descriptions, measures, hidden fields
Compares the model with a best-practice checklist.
Read ·
2. Evaluate model healthPoints to one of 5 reference files
STAR-SCHEMA, RELATIONSHIPS, MEASURES-DAX, PERFORMANCE and RLS.
Read ·
3. Targeted guidanceAdds DAX measures with format string and description
measure_operations with operation Create or Update.
Write ·
Create and update measuresDocuments columns and hides technical keys
column_operations with operation Update.
Write ·
Update column descriptions and hidden stateSets cardinality and cross-filter direction
relationship_operations with operation Create; the example uses OneDirection.
Write ·
Create relationshipsLists, creates and adds table permissions to roles
security_role_operations, shown in references/RLS.md.
Write ·
Row-level security rolesExecutes and validates DAX against the model
dax_query_operations is listed in the MCP tools reference.
Runs code ·
DAX queriesSearches Microsoft docs for current guidance
Uses microsoft_docs_search from the optional Microsoft Learn MCP server.
Network ·
Microsoft Learn lookups (optional)
From the official docs
Permissions
High risk
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only modeHow to turn it on
Watch out for
It can change your semantic model
SKILL.md's examples create measures and relationships and update columns and security roles. The MCP README says to create a backup before any operation and warns an LLM may make unintended changes. github.com
Model data can reach the AI provider
The MCP README says metadata, schemas or query results go to your MCP client and may be forwarded to the LLM provider. Avoid sharing chat sessions that include sensitive model content. github.com
You must accept a EULA for the npm install
The MCP README says the server blocks other tool calls until you accept its EULA. Read it before you use --accepteula or the environment variable for unattended runs. github.com
It reaches Fabric with your permissions
The README says the server connects with your existing credentials and Fabric RBAC. Use a least-privilege role. github.com
Security advisories
No published security advisories as of Nov 2, 2026.
Technical details
| Turning on read-only | Add --readonly to the MCP server args. Per the README it prevents any write operations to your semantic model. |
|---|---|
| Talks to | Fabric workspaces when you connect to one. Microsoft Learn (only if you add the optional Microsoft Learn MCP server) |
| Sign-in and keys | The MCP README says credentials are handled through the Azure Identity SDK and the server never stores or manages tokens directly. |
| Scopes |
|
Questions
What does the powerbi-modeling skill do?
It makes Claude act as a Power BI semantic modeling assistant. Claude connects to your open model, checks it against best practices and can create measures, relationships, descriptions and security roles.
Does it work on its own?
No. SKILL.md says the Power BI Modeling MCP server is required to connect to and change models. Without it Claude has no model to read. The install box includes the server setup from Microsoft's README.
Is this a Microsoft skill?
The skill is in GitHub's community awesome-copilot repo, and SKILL.md names no author. The MCP server it drives is Microsoft's: see the Power BI Authoring MCP server page.
Can it change or delete my model?
Yes. The skill's examples create and update measures, columns, relationships and security roles. Microsoft's README says to back up your model first, and the server has a --readonly option that blocks writes.
Does my data go to the AI provider?
It can. Microsoft's README says data the server retrieves, such as metadata, schemas or query results, goes to your MCP client and may be forwarded to the language model provider.
How do I install it?
Use the install box at the top of this page. It installs the skill and shows the MCP server config from Microsoft's README. The server's npm package needs Node.js and, per the README, you must accept its EULA before tools work.
Is it safe?
The skill bundles no scripts, but it drives a server that can edit your semantic model and run DAX queries. Microsoft's README says to back up your model first. Our tip: start the server in read-only mode for review work.
Information
updated Nov 2, 2026
- Repository
- github/awesome-copilot
- Folder
skills/powerbi-modeling
Sources
Docs, issue trackers and discussions this page is built from.
Show all 5Hide
- powerbi-modeling SKILL.md · github.com
- RLS.md · github.com
- PERFORMANCE.md · github.com
- Power BI Authoring MCP README · github.com
- awesome-copilot README · github.com