[chaitanyagiri]

What & why

Morning run, day 46 (12 Nov 2026): the daily brief, a launch page, a competitor page and a launch fix.

Merge order: #839, then #840, then #841, then #843, then this PR. This branch starts from #843's head (seo-engine/day-45-pm), with base main.

• New: agent-tools-today-2026-11-12 (1,098 words, 1 scene). Top 3: AWS Emulator (Amazon's 11 Nov post), the Gemini Decisions API (Google, 11 Nov) and the agents and lidar Maya site post (talk). The scene is a periscope; tapping a window turns it there and shows the date.
• New, launch slot: gemini-decisions-api-vs-openai-decisions-api (1,147 words, 3 scenes and a title card: prism, ballot, tollbooth). Google launched the Gemini Decisions API on 11 Nov at 18:00 UTC; we already had the OpenAI page, so this is a versus page with dated price, inputs and limits for both.
• New: marblism-alternatives (1,156 words, 3 scenes and a title card: tangram, waterwheel, hopscotch). A competitor post (High threat row): Sintra, Munder Difflin, Lindy, Grok Bot and Claude Cowork.
• Fix, launch desk: grok-bot-alternatives now covers Grok Bot 1.0 and xAI's new plans (Premium, Super and Ultra on The X Subscription, legacy Cursor plans) and corrects who picks the model, per xAI's security FAQ (1,196 words).
• Links: one sentence each in openai-decisions-api (to the Gemini page; its old house line is replaced with the current one) and claude-cowork-alternatives (to Marblism alternatives).

Type of change

• Bug fix
• New feature
• Refactor / cleanup
• Docs
• Build / CI

Checks

• gate.mjs (full, with media and motion): passed for all 6 touched posts.
• Fact review (two independent reviewers, every source fetched live):
  • Brief: FAIL once on 4 points, all fixed with the reviewer's text. Grok Bot 1.0 was dropped from the brief: xAI's page prints "Nov 2, 2026", and nothing dated it inside the window.
  • Grok Bot fix: FAIL once on 5 points, all fixed with the reviewer's text (who picks the model, the plans line, "out of beta" quoted from xAI's Grok Bot page, the pricing page link).
  • Gemini vs OpenAI: FAIL once on 3 small points (a quote, a caption source, a number in an alt), all fixed. Every fact matched Google's and OpenAI's pages.
  • Marblism alternatives: FAIL once on 6 points, all fixed (Lindy shows two prices, xAI's Premium plan, sale prices for Sintra, "businesses" as Marblism says, an opinion marked, an unquoted line removed).
  • Links: FAIL once on 2 points, both fixed.
• Scenes: each builder checked the frames at 25%, 50% and 75% and the stills; no Math.random; I checked the rendered pages.
• npm run build: passed (480 files written).

Notes

• OpenSEO is still signed out, so there was no Search Console pull, index watch or demand check this run.
• Queued from this run: openai-decisions-api (the API went generally available on 22 Oct; the page still says beta) and grok-bot-pricing (new plans).

Evidence

Images are on branch evidence/day-46-am.

Before

The blog hub and the three older pages this PR changes, on #843's head.

After

The blog hub, the three new pages, the three changed pages and the scene stills.

🤖 Generated with Claude Code