Snapshot of 17 Nov 2026, 18:00 Vienna time, proof of concept, version 3 (same data as version 1 and version 2; what changed).

What changed

Version 3 is the same frozen snapshot as version 1 and version 2, rebuilt after the developer manual and the sports coverage addendum. Any difference between the three links is these changes, not new data. The scoring formula itself is unchanged: alternatives will be compared side by side later.

Version 3

  1. My Sportlightname

    The name and the line "Your spotlight for sporting highlights", on every page, link preview and calendar file.

  2. No betting in publicmanual §08, §22

    No odds, no bookmaker names, no percentages anywhere on the site. Market prices stay an internal signal of how open a contest is. The expected outcome now comes from the table, form or rankings, and says so.

  3. Stories, not checklistsmanual §09 to §12

    A headline, a one-line hook on the card, about 110 words built on the strongest two to four angles and ending on an honest open question, and up to three things to watch for. Every sentence carries its own facts and passes both checks. The six checks from the first brief are now the review list beside each story.

  4. Richer factsmanual §10

    The last five results with scores, season leaders and the dated injury report for about 200 events, from data already in the snapshot. It also caught a v2 mistake: v2 billed a game as a superstar duel while one of the two stars was listed out injured. A star listed out no longer counts as taking part.

  5. Follows reorder, never rescoremanual §07

    A team, player or sport you follow moves its events up your lists. The score on the card stays the event's own.

  6. The bout is the unitaddendum §01 to §07

    The UFC card is split into its 6 main-card bouts, each with its own score, story and page, and at most 2 from one card in the top 10. Only the main card's start is known, so each bout says "bout time to be confirmed" and suggests watching from the main card start.

Version 2, from the first brief

  1. Six-check briefingbrief §8

    Each AI write-up answers why it matters, rivalry and recent history, current situation, who to watch, recent head-to-head and expected outcome, in 50 to 200 words. A check with no fact behind it says "No previous meeting in our sources" or "Not applicable" and is never filled in. The outcome is favoured, close contest or upset potential (in v2 set from the odds; since v3 from the table and form).

  2. Claim-level fact checkbrief §8, §14

    v1 checked that each cited fact exists. v2 also checks that every number, name, place and day is in a fact the sentence cites, that each number sits next to the side its fact is about, and then a second AI pass reads each check against only its cited facts. A write-up that fails twice is not published. This run: 21 events sent, 7 passed first time, 14 on the second draft, 0 not published; the checks caught a grounding problem in 13 of 21.

  3. Audit record per eventbrief §10

    Every briefing stores the model, the prompt version, the fact pack exactly as sent, the sources with when they were fetched, both checks' results for each draft, and two hashes. Each event page shows it under "Sources and checks". This is the source trail and audit log, without a CMS.

  4. Watchability, kept apart from importancebrief §7, §15, §22

    Its own component (10% of the base, the other five scaled to make room) and its own badge on every card: local start time over the event's length, and whether a viewing source is known. A 03:00 start reads "Middle of the night". The weights are a choice, not a finding, and stay in one config file.

  5. Hidden gem and varietybrief §7, §22

    One "Hidden gem" slot per day: the strongest event in a sport that is not big in your region, that you can watch at a sane hour. And no sport takes more than 3 of the top 10 tonight.

  6. Where to watch, only where a source saysbrief §11

    ESPN's schedule names US providers for 118 events; those show for US viewers. For Austria, the UK and the Netherlands no free source names a provider, so the page says "Viewing not verified" and nothing else: no guessed links, never an unofficial stream.

  7. Acceptance testsbrief §14

    Vienna and London times across a clock change; a postponed or cancelled event drops out of the lists and its page says so; no briefing is published without passing both checks; no viewing line for a region without a verified source.

Does the fact check bite?

A passing run proves nothing on its own, so each published briefing was corrupted on purpose and run through the checks again.

CorruptionCaught
One number changed to one no fact holds21 of 21
A number given to the other side (the fact is real, the team is wrong)15 of 15 (1 by the AI check)

Not yet, and why

From the briefWhy not now
Accounts, saved events, reminders, newsletterThey need a product with users. After the proof of concept shows the ranking is worth opening daily.
Human review of every eventThe aim is as much automation as possible: v2 reviews by exception (the fact checks, the audit record, write-ups that fail are not published). Whether the brand wants editors is an open question.
Full CMSThe audit record and the config files cover a proof of concept. A CMS comes with editors and a review queue.
Sportsbook odds comparison and affiliate linksNo betting in public for the first months; later only with legal advice per country and licensed feeds. Market prices stay an internal signal.
The full sports register and every sportThe catalogue is a decision list first: which sports to keep. The demo shows what free sources return for one week.
Editorial CMS and review queueThe aim is AI-only publishing with spot checks and feedback until the stories are right. The audit record on every page is the trail.
Affiliate and outbound-click trackingNo programme exists yet. Add it with the first real partner.
Creator pages, SEO landing pages, B2B widgetGrowth work, for once the product is worth growing.
Insider significance as its own score dimensionNeeds a source that is not popularity. Curated lists could stand in; the real version waits for specialist input.
Live data, simulated stakes, backtestStill next in line from v1. The backtest is what makes the weights defensible.
Native apps, 30+ sports, more languagesOn the brief's own later list.

21 events carry an AI briefing in this build. Every number on this page is read from the build itself. How the score works.