[Farzad Sunavala (farzad528)]

Description

Add Build a Microsoft IQ agent with the Microsoft Copilot harness in Foundry, a six-step walkthrough that deploys a reusable investigation skill, combines Fabric IQ records with a Foundry IQ account report, and inspects the resulting brief and tool trace.

Includes the registry entry, synthetic Adatum fixtures, sanitized live-run provenance, and 19 focused regression tests. Raw SDK captures and private environment configuration are not committed. No site implementation changes.

No linked issue.

Review focus

• Product/API accuracy and the distinction between Foundry IQ's shared project identity and Fabric IQ's delegated user identity.
• The restricted evaluation PDF/knowledge-base prerequisite is disclosed before setup. The local report transcription is not presented as an automatically configured retrieval source.
• SDK retrieval metadata and PDF text were verified. Native Playground report-citation click-through remains a manual review item and is not claimed verified. A skill guides the model; it does not enforce an application-level tool dependency graph.
• live-evidence.json separates historical captures from the exact supplied-copy revision tested at 2193d997d6b68926193606e06ab16018b70275bc. The retained cropped Playground screenshot is historical and is not embedded in the current walkthrough.

Validation evidence

Command / check	Result
python .agents\skills\cookbook-review\scripts\recipe-health-check.py notebooks\copilot-harness-foundry-fabric-iq.ipynb --repo-root .	Passed all 8 mechanical checks; advisory warning for intentionally cleared outputs.
python scripts\test-copilot-harness-recipe.py	Passed all 19 tests.
Set-Location 'scripts'; npx tsx validate-registry.ts	Passed all 20 registry entries; advisory long-title warnings only.
Set-Location 'site'; npm run build	Passed, including astro check: 0 errors, 0 warnings, 12 existing hints. npx astro check was not run separately.
git diff --check	Passed.
Clean-clone live execution of the recorded revision	All 11 unchanged code cells and the optional recovery snippet completed; one agent invocation, with no saved-response replay.
Independent source audit	50 supported checks; all 22 returned references matched 10 freshly downloaded PDFs; 283 telemetry rows recomputed.

The rendered notebook preview was checked locally. The Playwright suite was not run; this PR does not change site implementation. The author has confirmed that the three author-approval, internal-authoring-workflow, and human product-review requirements below have been met; these confirmations are not inferred from automated checks.

---

Checklist

Notebook contributions

Skip if this PR only touches site code.

• Recipe author is an approved Microsoft Forgebook author, or this contribution has maintainer approval
• Recipe was drafted or validated through the internal authoring workflow
• Foundry PM, engineering, or DX reviewer has checked product accuracy and positioning
• Notebook added to notebooks/ directory
• Entry added to registry.yaml with all required fields (slug, path, title, authors)
• Approved author entry exists in authors.yaml (with at least name and title)
• Tags are from the allowed list in tags.yaml
• Registry validation passes: cd scripts && npx tsx validate-registry.ts
• Notebook runs cleanly end-to-end
• Cell outputs cleared before committing
• No site-specific metadata inside the notebook
• No secrets, tenant-specific values, or private endpoints are committed
• Images stored in notebooks/media/<slug>/ with descriptive alt text
• Data files stored in notebooks/data/<slug>/ when applicable

Site changes

Skip if this PR only adds a notebook.

• Site builds without errors: cd site && npm run build
• No TypeScript errors: cd site && npx astro check

[github-actions]

🔍 Preview

Build and checks passed, but a live preview wasn't auto-deployed because Farzad Sunavala (@farzad528) isn't a member of forgebook-contributors yet. A maintainer can add you to the team and re-run this workflow to deploy the preview.