ABC Corp is a fictional company. Every name, number and date is invented. This is a reference artifact generated with an LLM coding agent; the brief that produces it is at the bottom of this page.
| Severity | Risk | Owner | Mitigation |
|---|---|---|---|
| HIGH | Duplicate customer records in wave-2 source data could corrupt billing history on migration. | S. de Wit | Dedupe script run before each batch; manual review of flagged records; batch 3 gated on a clean report. |
| MED | Vendor API rate limits during peak migration windows may stretch the wave-2 schedule. | R. Okafor | Night-window batching agreed with vendor; throughput test scheduled 16 Jun. |
| MED | Key subject-matter experts on leave through July, so decommission review capacity is thin. | A. Kowalczyk | Review sessions front-loaded to June; deputy reviewers named for each area. |
Approve front-loading decommission reviews into June so the sign-off can hold at 17 Jul. Without sign-off cover, the 2-week go-live buffer is at risk. Owner: A. Kowalczyk · needed by Fri 19 Jun.
Confidence in the 28 Aug go-live: MEDIUM — held on the current plan, but contingent on a clean wave-2 data-quality run and change-board approval first pass.
Here are my raw status notes for this week: [paste your notes]. Create a complete single-file HTML status one-pager for the project: overall RAG status with a one-line reason, a milestone timeline showing done / in progress / at risk, budget spent vs plan as a simple bar, top 3 risks with owners and mitigations, what changed since last week, and next week's focus. Use only inline CSS and SVG. No external libraries, no images. It must print cleanly on one A4 page.
Paste the brief into any capable LLM: GPT, Claude, Gemini, Grok, DeepSeek, or the assistant your company
provides. Iterate a few rounds on layout and content until it reads well. Save the final answer as a
.html file and open it in any browser. Expect similar output, not identical: every model has its
own taste, and that is fine.
This reference artifact was built with Claude Code, an LLM coding agent, over several iterations. Treat it as the bar to aim for, not as a guaranteed first answer. All data on this page is fictional.