REPO PACKAGE
microsoft semantic-link-labs
OFFICIAL13skills
4ready to use
LABSCO SUMMARY
Every one of the 13 is a guide to working on the library's own source, not a wrapper around its Fabric or Power BI functions: add-function and rest-api-patterns cover extending the library's own API-wrapper pattern, direct-lake-operations and tom-operations and sql_to_dax cover the parts that touch Direct Lake, the Tabular Object Model, and DAX specifically, and the rest — build-docs, code-style, run-tests, write-tests, pr-review-comments, github-repo-explore, planning-with-files, ui-styling — are generic open-source contributor practice that would fit almost any Python repository with a docs site, a linter, and a pytest suite.
This is for someone opening a pull request against microsoft/semantic-link-labs itself — extending a wrapper function, fixing a DAX translation, reviewing someone else's PR — not for a Fabric or Power BI user who wants an agent to run the library's actual features (migrating a semantic model, analyzing a report, managing a capacity) against their own workspace. None of the 13 skills mention sempy_labs' own functions, workspaces, or a live Fabric connection; they are about the repository's engineering process, not its product.
READ THE FULL ANALYSIS
What the dependency labels say. Our check couldn't confirm what five of the thirteen actually need to run and marked them account_key, filed five more as local_tool, and left only three as ready to use as-is — which tracks with a dev-workflow pack: several assume a working repo checkout, a GitHub token for gh, and a Python test environment already in place, none of which the skills themselves set up.
Checked 19 September 2026 against microsoft/semantic-link-labs' own skill files and README; the contributor-versus-user split above is our reading of each skill's description, not a distinction Microsoft states itself.
WHAT'S INSIDE
13 showing · 13 totalReady to use
Nothing else to set up — install it and go.
13 of 13
SETUP
Showing all 13 skills
add-function
A contributor's walkthrough for getting a new feature into Semantic Link Labs, Microsoft's Python library for Power BI and Fabric — which folder the code belongs in, how it has to be written, and the checklist it must pass before it is submitted.
build-docs
Turn the project's code comments into its documentation website on your own machine, so formatting mistakes and gaps show up before readers see them.
code-style
Run the project's formatter, style checker and type checker over the Python code, and fix what they complain about before committing.
direct-lake-operations
Power BI models that read straight from a data lake instead of importing a copy: how to build them, keep them in step with the lake's tables, and work out why one has quietly gone back to the slow way.
github-repo-explore
Find an open-source project on GitHub, copy it down to a folder on your machine, and read how it solved the problem you are working on.
planning-with-files
A way of keeping a long job on track: the plan, what has been found and what is finished all go into files on disk, so nothing is lost when the assistant's memory fills up.
pr-review-comments
Leave code-review notes attached to the exact lines of code they are about in a GitHub pull request, instead of one lump comment at the bottom.
rest-api-patterns
This library reaches Microsoft's Fabric and Power BI web services through one shared piece of code; here is how to use it, and what to do when a call is paged, slow, or fails.
run-tests
Run the project's automated tests on your own machine — all of them, or just the few you care about — and dig into the ones that fail.
sql_to_dax
Rewrite a calculation you already have in SQL as the Power BI formula that means exactly the same thing — not one that merely looks similar.
tom-operations
A Power BI data model opened as Python objects: its tables, columns, measures and relationships can be listed and edited in code instead of by hand in the desktop app.
ui-styling
The shared look and building blocks for the interactive panels this library draws inside a notebook, so a new one matches the ones already there instead of inventing its own style.
write-tests
New code here needs tests that cover the ordinary case, the awkward inputs and the way the function fails — these are the patterns the project uses to write them.
No skill in this package matches that.
HOW TO GET IT
WHOLE PACKAGEInstall the whole package (13 skills)
npx skills add microsoft/semantic-link-labsSINGLE SKILLOr install a single skill
npx skills add microsoft/semantic-link-labs --skill <name> --full-depthPick the skill name from the Skills tab — each entry there installs independently.