This AI Agent Builds $15K Cinematic Websites on Autopilot (Claude Code + Nanobanana 2)
This video shows a 'Cinematic Sites' agent skill — a single skill.md run by Claude Code (here via a Telegram-connected session) — that takes an existing website and runs a four-step pipeline: analyze the brand, generate cinematic hero scenes with Nano Banana images animated into video via Kling, build a scroll-driven website, and deploy to Vercel.
Jay E | RoboNuggetsWatchTranscript found
Quick learning frame
Read this before watching.
Coding-agent workflow is the loop of inspect, plan, edit, verify, summarize, and route the next task to the right tool.
New playlist item from Jay E | RoboNuggets; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to encode a multi-step creative web-build pipeline into a single agent skill.md that orchestrates brand analysis, image/video generation, scroll-frame website assembly, and deployment with human approval pauses.
Watch for the shift from claim to mechanism. The learning value is the point where the transcript reveals a repeatable action, tool boundary, context move, review habit, or artifact.
Concept diagram
Where this video fits.
01Inspect
02Plan
03Edit
04Verify
05Review
06Route
Deep lesson
Turn this video into working knowledge.
3,746 cleaned transcript words reviewed across 1,036 timed caption segments.
Thesis
This AI Agent Builds $15K Cinematic Websites on Autopilot (Claude Code + Nanobanana 2) teaches a practical codex + claude workflows move: This video shows a 'Cinematic Sites' agent skill — a single skill.md run by Claude Code (here via a Telegram-connected session) — that takes an existing website and runs a four-step pipeline: analyze the brand, generate cinematic hero scenes with Nano Banana images animated into video via Kling, build a scroll-driven website, and deploy to Vercel.
The goal is not to remember the video. The goal is to extract the operating principle, tie it to timestamped evidence, test how far the claim transfers, and make something reusable.
0:00
Skill-driven pipeline
“I just created an agent that auto builds cinematic responsive websites from scratch. You drop in your current website, it analyzes your brand's color palette, your company's details, and designs a new website for you. Using this, I...”
The whole agent kit is one skill.md — an instruction manual telling the agent to run four steps (analyze brand, generate scene, build site, deploy) — and step one needs no external tools since agentic platforms like Claude Code can read HTML directly (Firecrawl optional); it pauses to let you approve a generated brand card capturing industry, color palette, and typography. Write the skeleton of a skill.md that lists the four pipeline steps as plain-English instructions and add an explicit human approval pause point after the brand-analysis step.
5:45
Images then video
“Agentyc platforms like CloudCode or Antigravity or even Open Claw can read HTML files, then if you provide the website that you want to extract these components for, they will be able to do a pretty good job...”
Scene generation has the agent suggest ~3 hero concepts, generate images with Google's Nano Banana (using the $300 welcome credit per Gmail account, no third-party reseller needed), then animate the chosen image into video via Kling V3 image-to-video accessed through the Wave Speed aggregator; doing two video variants raises odds given the probabilistic output. Grab a Nano Banana documentation link and a Wave Speed model page, copy that documentation to your agent, and have it generate one hero image then animate it into two video variants so you can pick the steadier one.
10:27
Scroll-frame build
“agent with the knowledge to understand how this works, all you need to do with a lot of these websites like Wave Speed, key.ai, is that you just need to copy the content here. And let's say you...”
The build step codifies the cinematic effect by extracting individual video frames and mapping them to scroll position, and pulls from a 31-piece cinematic modules pack (accordion sliders, reveal text, kinetic text, image trails, SVG draw animations) so the agent can drop in professional responsive components on request. Take a generated site and prompt the agent to swap its menu into the accordion-slider cinematic module, then verify the hover-expand behavior renders as intended.
01
Inspect
Start with this video's job: This video shows a 'Cinematic Sites' agent skill — a single skill.md run by Claude Code (here via a Telegram-connected session) — that takes an existing website and runs a four-step pipeline: analyze the brand, generate cinematic hero scenes with Nano Banana images animated into video via Kling, build a scroll-driven website, and deploy to Vercel. Treat "Inspect" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:00, where the video says: “I just created an agent that auto builds cinematic responsive websites from scratch. You drop in your current website, it analyzes your brand's color palette, your company's details, and designs a new website for you. Using this, I...”
02
Plan
Use "Plan" to locate the part of the codex + claude workflows workflow the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 5:45, where the video says: “Agentyc platforms like CloudCode or Antigravity or even Open Claw can read HTML files, then if you provide the website that you want to extract these components for, they will be able to do a pretty good job...”
03
Edit
Turn "Edit" into the reusable artifact for this lesson: A routing matrix for when to use Codex, Claude, browser checks, or manual review. This is where watching becomes something you can inspect and reuse.
04
Verify
Use "Verify" as the application surface. Decide whether the idea touches a browser flow, a local file, a model choice, a source document, a UI, or a review step.
05
Review
Use "Review" to prove the lesson. The evidence should connect back to the video title, transcript anchors, and a concrete output, not a generic best-practice claim.
06
Route
Use "Route" to carry the idea forward: save the prompt, checklist, diagram, or operating rule that would make the next agent run better.
Example
Source-backed work packet
Convert the video into a scoped task that includes the transcript claim, target workflow, acceptance criteria, and proof. The output should be a routing matrix for when to use codex, claude, browser checks, or manual review..
Example
Claim vs. demo brief
Separate what the speaker claims, what the demo actually proves, and what still needs outside verification before you adopt the workflow.
Example
Teach-back module
Transform the lesson into a definition, a mechanism diagram, one misconception, one practice exercise, and a check-for-understanding question.
Do not learn it wrong
Treating the title as the lesson without checking what the transcript actually says.
Letting the prompt drift into generic advice that could apply to any video in the playlist.
Copying the tool setup without identifying the operating principle that transfers to your own stack.
Skipping the artifact, which means the learning never becomes operational or inspectable.
Do not count this as learned until these are true.
01
State the transcript-backed claim in your own words: This video shows a 'Cinematic Sites' agent skill — a single skill.md run by Claude Code (here via a Telegram-connected session) — that takes an existing website and runs a four-step pipeline: analyze the brand, generate cinematic hero scenes with Nano Banana images animated into video via Kling, build a scroll-driven website, and deploy to Vercel.
02
Explain the practical stakes without hype: New playlist item from Jay E | RoboNuggets; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the Inspect -> Plan -> Edit -> Verify -> Review -> Route sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A routing matrix for when to use Codex, Claude, browser checks, or manual review.
Put it into practice
Give this grounded prompt to Codex or Claude after watching.
You are helping me turn one specific YouTube video into real, durable learning.
Source video:
- Title: This AI Agent Builds $15K Cinematic Websites on Autopilot (Claude Code + Nanobanana 2)
- URL: https://www.youtube.com/watch?v=bUt1WpDlI6E
- Topic: Codex + Claude Workflows
- My current learning frame: Build a minimal skill.md that takes one existing website, generates a brand card and a Nano-Banana-to-Kling cinematic hero, assembles a scroll-frame page using one cinematic module, and deploys it to Vercel as a shareable demo.
- Why this matters: New playlist item from Jay E | RoboNuggets; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:00 / Evidence 1: "I just created an agent that auto builds cinematic responsive websites from scratch. You drop in your current website, it analyzes your brand's color palette, your company's details, and designs a new website for you. Using this, I..."
- 3:12 / Evidence 2: "is accessible anywhere using this tool called Vercel. And like I mentioned earlier, you can do this on Open Claw, on CloudCode, or whatever Agentyc platform you're using. For what I'll demo now, I'll do it via this..."
- 5:45 / Evidence 3: "Agentyc platforms like CloudCode or Antigravity or even Open Claw can read HTML files, then if you provide the website that you want to extract these components for, they will be able to do a pretty good job..."
- 7:27 / Evidence 4: "their Nano Banana or ViO models. So, when you set up this skill for the first time, that will also show you where to find your API keys that you can just provide your agent so that they're..."
- 10:27 / Evidence 5: "agent with the knowledge to understand how this works, all you need to do with a lot of these websites like Wave Speed, key.ai, is that you just need to copy the content here. And let's say you..."
- 12:27 / Evidence 6: "description below. But this just gives you a lot of modular elements to enrich any website build that you have for your agents. So, there's this one with a reveal text effect, modular components around future scrolling. This..."
- 14:43 / Evidence 7: "going to do is to just deploy that to Vercel. And basically, Vercel is a service that hosts websites for you. And they actually do this for free, which is really good. And for you to equip your..."
Your task:
1. Use the transcript anchors above as the primary source packet. If you add outside context, label it clearly as outside context and keep it secondary.
2. Create a source-check table with columns: timestamp, claim, what the demo proves, confidence, and what still needs verification.
3. Extract the actual teachable claims from the video. Do not invent claims that are not supported by the title, lesson frame, or transcript anchors.
4. Build a reusable learning artifact: A routing matrix for when to use Codex, Claude, browser checks, or manual review.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: Inspect -> Plan -> Edit -> Verify -> Review -> Route
- 3 concrete examples that apply the video idea to real agentic work
- 2 failure modes the video helps prevent
- a checklist I can use the next time I run Codex or Claude
- one practical exercise with a clear done signal
6. Add a "learning transfer" section: what changes in my workflow tomorrow if I actually learned this?
7. Add a "source check" section that cites which transcript anchor supports each major takeaway.
Quality bar:
- Make this specific to "This AI Agent Builds $15K Cinematic Websites on Autopilot (Claude Code + Nanobanana 2)", not a generic Codex + Claude Workflows essay.
- Prefer operational examples, failure modes, and reusable artifacts over broad definitions.
- Call out uncertainty instead of smoothing over weak evidence.
- If evidence is weak, say what transcript segment or timestamp needs review instead of guessing.
- Finish with a concise artifact I could paste into my learning app.
Misconceptions
What to stop believing.
One agent should do every task.
Different tools have different strengths. Routing is part of the workflow.
More context is always better.
Relevant context helps; stale context causes drift and cost.
Practice studio
Learning only counts when you make something.
01
Transcript evidence map
Separate what the video actually says from what you already believe about the topic.
3 source-backed takeaways with timestamps, confidence, and a transfer note.02
One useful artifact
Apply the video to a real workflow and produce a routing matrix for when to use codex, claude, browser checks, or manual review..
A reusable artifact with a done signal and one verification step.03
Teach-back card
Explain the lesson to someone who has not watched the video yet.
A 90-second explanation, one diagram, one example, and one misconception to avoid.
Recall check
Answer first, then reveal — without rewatching.
What is the 'agent kit' that runs the whole cinematic-sites pipeline actually made of, what four steps does it drive, and why does step one need no external scraping tool?
In the scene-generation step, which two models do images and video, how is the image model accessed cheaply, and why does the presenter generate two video variants?
How does the build step actually create the 'cinematic' scroll effect, and what reusable resource lets the agent drop in professional components like the menu accordion?
Source shelf
Use the video as a doorway, then verify with primary sources.
