docs(ci-readiness-check): require explicit mode prompt before running

[noencke]

Description

A user reported that the ci-readiness-check skill would silently run in
Build mode (the expensive one) every time they pushed, instead of asking
which mode to run. Step 1 of the skill already said to ask the user, but
the directive was buried below the <required> block, soft-worded
("Step 1 asks the user..."), and not enforced against auto/autonomous mode.

This change rewrites the lead of the <required> block to make the prompt
unmissable:

• States as a MUST that the agent ask the Step 1 mode-choice question first.
• Explicitly disqualifies "we're in auto mode" as a justification to skip.
• Defines exactly two narrow exceptions where the agent may proceed without
  asking (no-op since the previous check in this conversation; explicit
  standing user instruction), with conservative wording so subjective "I
  think nothing important changed" does not qualify.
• Requires that, in both exceptions, the agent still report the skip and
  which exception applied — every time, with no abbreviation across
  repeated runs.

Reviewer Guidance

The review process is outlined on this wiki page.

[github-actions]

Hi! Thank you for opening this PR. Want me to review it?

Based on the diff (11 lines, 1 files), I've queued these reviewers:

• Correctness — logic errors, race conditions, lifecycle issues
• Security — vulnerabilities, secret exposure, injection
• API Compatibility — breaking changes, release tags, type design
• Performance — algorithmic regressions, memory leaks
• Testing — coverage gaps, hollow tests

Toggle the reviewer checkboxes above to adjust, then tick the box below to start:

• Start review

[Copilot]

Pull request overview

Updates the ci-readiness-check skill’s <required> preamble to prevent silent “Build mode” runs by requiring an explicit user mode choice up front, even in auto/autonomous operation.

Changes:

• Makes the Step 1 mode-selection prompt an explicit MUST-do-first instruction.
• Defines two narrow exceptions for when the prompt may be skipped and requires explicit reporting when an exception is used.
• Reiterates task/todo creation requirements immediately after the user responds.

[github-actions]

🔗 No broken links found! ✅

Your attention to detail is admirable.

linkcheck output

> fluid-framework-docs-site@0.0.0 ci:check-links /home/runner/work/FluidFramework/FluidFramework/docs
> start-server-and-test "npm run serve -- --no-open" 3000 check-links

1: starting server using command "npm run serve -- --no-open"
and when url "[ 'http://127.0.0.1:3000' ]" is responding with HTTP status code 200
running tests using command "npm run check-links"


> fluid-framework-docs-site@0.0.0 serve
> docusaurus serve --no-open

[SUCCESS] Serving "build" directory at: http://localhost:3000/

> fluid-framework-docs-site@0.0.0 check-links
> linkcheck http://localhost:3000 --skip-file skipped-urls.txt

Crawling...

Stats:
  288641 links
    1922 destination URLs
    2172 URLs ignored
       0 warnings
       0 errors