chore: weekly message center posts refresh 2026-07-27

[soyalejolopez]

Recovers four weeks of Message Center refreshes that were generated but never merged.

The weekly workflow pushed its branch successfully every week since 7/6, then died on
gh pr create because the org blocks GitHub Actions from opening pull requests. The
content was sitting on message-center/2026-07-06 through -07-27 the whole time —
no PR was ever created, so nobody saw it. copilot-agents-guide/index.html has been
stale since July 6.

Each weekly run regenerates the full messageCenterPosts array from the live feed, so
this branch supersedes all three earlier ones. Merging this alone recovers every week.
The other three can be deleted.

New in this refresh (10):

• MC1434583 — Copilot: configure org policy link for blocked Copilot Chat users
• MC1427972 — SharePoint: restricted content discovery for Copilot and M365 search
• MC1423101 — M365 Copilot usage reports: faster refresh, more Graph API metrics
• MC1422074 — OpenAI models as a subprocessor in Microsoft 365 Copilot
• MC1419800 — SharePoint button web part: custom Copilot prompts / Power Automate
• MC1419795 — PowerPoint web: one-click Review Presentation and Visualize Slide
• MC1418558 — Taskbar pinning behavior change for M365 Copilot and companion apps
• MC1413308 — Purview DLM: insights for Copilot and AI app interactions
• MC1413298 — Copilot Studio and agents moving to the Microsoft 365 Roadmap
• MC1411132 — Power Automate UI automation repair agent

Note for review: the post list goes from 22 entries to 14. Four are retained and 18
age out, which is the expected rolling-window behavior but is a larger turnover than a
normal week because four weeks landed at once. Entries also switch from JS object
literal syntax to JSON syntax — valid either way, no render change, but it widens the diff.

Once AUTOMATION_PAT is set, this workflow opens its own PR again and this manual step
goes away.

[github-actions]

🛰️ PR Sweeper report

Risk: 🟢 LOW · Security gate: ✅ passing · Files: 1

🔒 Automated guardrails (authoritative)

No secret, PII, file-policy, or scope issues detected. ✅

🤖 Dual-model AI review (advisory)

AI review unavailable for this run (models not reachable or no diff). Guardrails above are unaffected.

• GPT-class reviewer (openai/gpt-5): skipped — GitHub Models HTTP 413: {"error":{"code":"tokens_limit_reached","message":"Request body too large for gpt-5 model. Max size: 4000 tokens.","details":"Request body too large for gpt-5 model. Max size: 4000 tokens."}}
• Deep-reasoning reviewer (openai/o3): skipped — GitHub Models HTTP 413: {"error":{"code":"tokens_limit_reached","message":"Request body too large for o3 model. Max size: 4000 tokens.","details":"Request body too large for o3 model. Max size: 4000 tokens."}}

---

The automated guardrails are authoritative and gate the security status. The AI review is advisory and never auto-merges. Thanks for contributing to FastTrack! 🛩️