Copilot Licensing Advisor skill

[Tim (Timziito)]

Interview-driven recommendation of the right Microsoft 365 Copilot, Copilot Studio, and Power Platform license mix, with costs fetched live at the time of the conversation instead of hardcoded prices.

[Copilot]

Pull request overview

Adds a new copilot-licensing-advisor submission to the skills gallery, aimed at interviewing users and producing a license mix + cost table for Microsoft 365 Copilot, Copilot Studio, and Power Platform using live-fetched pricing at conversation time.

Changes:

• Added the agent-facing SKILL.md with an interview-first flow, guardrails against quoting unfetched pricing, and a cost-table output requirement.
• Added a references/sku-rules.md “starting checklist” for slower-moving eligibility rules (explicitly not pricing).
• Added human-facing README.md and submission metadata.json.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated no comments.

File	Description
submissions/copilot-licensing-advisor/SKILL.md	Defines the agent procedure (interview → fetch current info → map to SKUs → cost table) and guardrails.
submissions/copilot-licensing-advisor/references/sku-rules.md	Provides a baseline eligibility checklist to cross-check against fetched, current docs.
submissions/copilot-licensing-advisor/README.md	Explains skill intent and links to official references for humans browsing the gallery.
submissions/copilot-licensing-advisor/metadata.json	Registers the submission metadata (name/description/platforms/tags/author/version/dates).

[Adriana Trujillo (adrianatruji)]

Thanks for this, Tim (@Timziito) . Licensing is exactly the kind of high-stakes question where a good digest helps people, so I'm glad you're aiming at this one.

The thing I'd push on is that even the "structural, slow-changing eligibility rules" reflected in the sku-rules.md, will move eventually, and understand how they change is the kind of exercise that would require this type of assistance, so the file ships with contradictions because of it. The recently released Copilot Studio usage-based billing scheme is a good example: The sku-rules statements about Microsoft 365 Copilot license coverage for Copilot Studio (e.g. "Agent capability used inside Microsoft 365 Copilot … doesn't consume the separate Copilot Studio message/credit meter,") aren't accurate anymore under the new scheme. So the "stable" rules are already stale and the exercise of adjusting them is where AI assistance like this could be handy.

Instead of encoding licensing concepts as static rules that quietly go out of date, lean into what a skill is uniquely good at: Turning ambiguous or spread documentation into something easy to digest. A skill that builds a visual/interactive licensing decision guide from live-fetched, dated sources (and is honest about what it couldn't verify) would be more useful to the person on the other end. The 'interview' flow you already have is a good backbone for that.

Keen to review the changes that come from this, it's indeed a hot topic! Thanks again.