Add post: Evaluation-Driven Agent Readiness in Copilot Studio

[KarimaKT]

Here's a Draft for Adi. Screenshots and some tweaks missing.

Summary

• New maker-focused post: Evaluation-Driven Agent Readiness in Copilot Studio
• Walks through scoping a narrow agent, bucketing coverage, generating realistic test sets, stacking graders, reading failures, fixing the design, and rerunning to prove it
• Uses a Kcontoso shipping-support agent as the running example; links to MS Learn agent-evaluation + triage/remediation docs

Still to do

• Add the 8 screenshots (header + 01-07) under assets/posts/evaluation-driven-agent-readiness/
• Minor copy tweaks
• Flip published: false to true when ready

Checklist

• Ran /review-post and addressed feedback
• Local server renders correctly (./tools/run.sh)
• All images have alt text and captions
• 2-3 internal links to related posts
• Tags chosen for Chirpy "Further Reading" overlap

[github-actions]

Blog compiled, but preview generation failed for 356ec70.

The Jekyll build completed, but one or more HTML or screenshot previews could not be generated. See the workflow run or the result artifact.

Missing previews:

• \_posts/2026\-06\-01\-evaluation\-driven\-agent\-readiness\-copilot\-studio\.md: Expected generated page was not found at _site/mcscatblog/posts/evaluation-driven-agent-readiness-copilot-studio/index.html.