agentleFS
MicrosoftKnown publishermicrosoft/mcs-labs/.github/skills/lab-update/SKILL.md

lab-update

Skill152 starsChanged 7 days agoMITView on GitHub

Fix or refresh an existing MCS Labs hands-on lab (classic or guided missions format) and open a pull request. Applies findings from a lab-review report, a lab-runthrough log, a GitHub lab-update issue or a product change, keeps the README and _labs copies in sync, replaces drifted screenshots, keeps durations and agendas consistent, records confirmed product renames for the audit, validates, and opens a reviewable PR. Use when asked to fix, update, refresh, correct or modernize a lab, apply review findings, update screenshots, handle a product rename across labs, or open a PR for lab changes. Never merges.

  • Commits and pushes

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. id-andyyy/LabAI★ 3same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository7 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 6 are distinctive, and 6 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 6

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  • could be automatic: a setting could enforce it
  1. Distinctive:

    Every update is evidence, then a small coherent change to both copies, then validation, then a PR someone else merges.

    line 8
  2. Distinctive:

    Use one of these, in order of preference:

    line 13
  3. Distinctive:

    UI labels: Use the label exactly as the product shows it, in bold, and quote it in the PR.

    line 46
  4. Distinctive:

    Dates: Do not add update dates; the site derives them from Git.

    line 54
  5. Very specific:

    Missions labs: Keep the contract: core minutes sum to duration, each check has exactly one correct option, optional missions are marked, …

    line 55
  6. Very specific:

    When a rename is confirmed, add it to scripts/lab-audit-terms.json with term, pattern (JavaScript regular expression, case-insensitive), …

    line 64
  7. Very specific:

    Use the captures and crop sizes from the lab-runthrough log.

    line 72
  8. Very specific:

    Delete an image only when the audit reports it as image-orphan and nothing else references it (git grep <file name>).

    line 78
  9. Very specific:

    Commit both copies, images and any terms or agenda changes, with a message such as fix(<slug>): update Cowork plugin steps for the new …

    line 97
  10. Very specific:

    Check CI (gh pr checks <n>): Lab Doc Sync, Accessibility and Lighthouse.

    line 101
  11. Distinctive:

    Never merge your own PR, and never push to main.

    line 102
  12. Distinctive:

    When asked to keep labs current after a product release:

    line 106

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Update a lab

Every update is evidence, then a small coherent change to both copies, then validation, then a PR someone else merges.

1. Start from evidence

Use one of these, in order of preference:

  1. A lab-runthrough log: what really happened in the product.
  2. A lab-review report: audit findings, product claims checked against Microsoft Learn, learning-design scores.
  3. A GitHub issue (gh issue view <n> --repo microsoft/mcs-labs), usually from the Lab Update form.
  4. A product announcement. Confirm it on Microsoft Learn or the product itself before changing a lab; announcements often precede rollout.

With only a vague request ("this lab is out of date"), run lab-review first. If the lab depends on product UI and a training environment is available, run lab-runthrough too.

Agree the scope before editing when it is more than a fix: rewrites, a new format, different prompts or a new duration need the lab owner (git log --format='%an' -- labs/<slug> _labs/<slug>.md | sort | uniq -c | sort -rn).

2. Branch

Work on a feature branch, never on main. One PR per coherent change: one lab's refresh, or one rename across many labs. Do not mix a content refresh with unrelated layout or script changes.

3. Edit both copies

labs/<slug>/README.md (GitHub) and _labs/<slug>.md (site) change together, in the same commit. CI (Lab Doc Sync) fails a PR that changes a README without its site copy.

  • Headings. README headings keep their emoji; _labs headings have none. When a heading changes, update the hand-written table of contents in both copies; the audit checks every anchor.
  • Voice and scope. Fix what the evidence shows. Keep the lab's structure and voice; do not rewrite a working section to taste.
  • UI labels. Use the label exactly as the product shows it, in bold, and quote it in the PR. When a label alternates between rollouts, name the destination or give both.
  • Prompts. Prompts are tested content. Change one only when the run-through shows it no longer works, and say why in the PR. In missions labs the prompt must stay identical in front matter, the _labs body and the README.
  • Durations. duration, the Lab Details row, use-case or mission minutes, and every agenda slot listing the lab (agenda-usage) must agree. Agenda changes affect an event, so name the event in the PR.
  • Dates. Do not add update dates; the site derives them from Git.
  • Missions labs. Keep the contract: core minutes sum to duration, each check has exactly one correct option, optional missions are marked, and cleanup stops anything recurring. npm test enforces it.
  • Retired names. Current names (check scripts/lab-audit-terms.json for the full list): Microsoft Copilot and Microsoft Copilot Chat since August 2026 (formerly Microsoft 365 Copilot and Microsoft 365 Copilot Chat), and the Microsoft Copilot harness in Copilot Studio since October 2026 (formerly GitHub Copilot harness). Keep license SKU names, such as Microsoft 365 Copilot (add-on), as the price list shows them, and UI labels exactly as the product shows them. When a rename is confirmed, add it to scripts/lab-audit-terms.json with term, pattern (JavaScript regular expression, case-insensitive), replacement, since and an official Microsoft source URL. Then run npm run audit:labs -- --all --format markdown and list other affected labs as follow-ups, or fix them in a separate rename PR.

4. Replace screenshots

Use the captures and crop sizes from the lab-runthrough log. Keep the file name and format so neither copy's Markdown changes. For missions labs, update width, height and alt in the front matter, keep src and src_dark pairs together and keep the <picture> element in both bodies. Look at every new image before committing: subject legible, both themes framing the same content, no personal data outside the scenario.

Delete an image only when the audit reports it as image-orphan and nothing else references it (git grep <file name>). Shared images (image-shared) belong to more than one lab.

5. Validate

npm run audit:labs -- --changed   # no new errors or warnings in the labs you touched
npm test                          # not run by CI; run it locally
bash tools/run.sh                 # or .\tools\run.ps1; open /mcs-labs/labs/<slug>/

In the browser check the changed sections, images in light and dark themes, the table of contents links, and Previous and Next with ?event=<id> for each event that lists the lab. Compare the audit with main when the lab already had findings: the PR should not add any.

6. Open the PR

  1. Commit both copies, images and any terms or agenda changes, with a message such as fix(<slug>): update Cowork plugin steps for the new Agents & Plugins page.
  2. Push the branch and open the PR against microsoft/mcs-labs main with gh pr create and a body from templates/pr-body.md. Link the issue with Fixes #<n>.
  3. Check CI (gh pr checks <n>): Lab Doc Sync, Accessibility and Lighthouse.
  4. Never merge your own PR, and never push to main.

Recurring maintenance

When asked to keep labs current after a product release:

  1. npm run audit:labs -- --all --format markdown for the static picture.
  2. lab-review on labs about the changed product, checking claims on Microsoft Learn.
  3. lab-runthrough on labs in upcoming events first (_data/agendas/*.yml).
  4. One PR per lab or per rename. Put pre-existing findings you did not fix under "Follow-ups" in the PR body rather than widening the PR.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.