This video shows how to set up the Hyperframes plugin inside OpenAI's Codex app so an AI agent can produce and edit videos by writing HTML code — the timeline itself becomes code — covering the new HTML-in-canvas feature, parallel threads via built-in git worktrees, and real builds like a liquid-glass subscribe animation, a website product demo, and an MP3 waveform visualizer.
David OndrejWatchTranscript found
Quick learning frame
Read this before watching.
Creative automation uses agents to accelerate production while keeping human taste in story, pacing, selection, and critique.
New playlist item from David Ondrej; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to configure Codex with the Hyperframes plugin and drive AI-generated motion graphics through plain-English prompts, iterating with parallel threads, pre-sent feedback, and token-saving habits like /compact.
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.
01Brief
02Source
03Generation
04Selection
05Edit
06Taste Review
Deep lesson
Turn this video into working knowledge.
4,336 cleaned transcript words reviewed across 1,354 timed caption segments.
Thesis
Codex can now make videos… it’s insane teaches a practical creative automation move: This video shows how to set up the Hyperframes plugin inside OpenAI's Codex app so an AI agent can produce and edit videos by writing HTML code — the timeline itself becomes code — covering the new HTML-in-canvas feature, parallel threads via built-in git worktrees, and real builds like a liquid-glass subscribe animation, a website product demo, and an MP3 waveform visualizer.
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.
0:58
Timeline becomes code
“Premiere, Final Cut, CapCut, these all require a human dragging clips on a timeline. Now, Hyperframes completely flips this because the timeline itself becomes code. And now, an AI agent like Codex, Hermes, Claude, they can produce, edit,...”
The bottleneck in editing was never rendering, it was the human dragging clips in Premiere or CapCut; Hyperframes flips this by making the timeline itself code, so agents like Codex, Hermes, or Claude can produce and manipulate videos by writing plain HTML — something they are already extremely good at. Install the official Hyperframes plugin in the Codex app (GPT 5.5, speed on, auto-review mode), open an empty project folder, and run one motion-graphic prompt end to end.
11:04
Parallel threads workflow
“sure to use {slash} compact a lot, okay? This is a built-in command into Codex that will compact the chat history to save you tokens because this can eat up your limits very fast. Now, obviously, Codex has...”
Because Codex has git worktrees built in, you can launch a second prompt (like a 3D liquid-glass subscribe animation) in a new thread while the first composition is still cooking, and small follow-up edits like a red-gradient recolor finish in about 2.5 minutes since the agent only rewrites the relevant CSS, not the whole build. Run two Hyperframes prompts in parallel threads — one new composition and one color-only tweak to an existing one — and note the time difference between a full build and a scoped edit.
16:18
Reusable asset library
“see, Codex just opens this. I didn't even do anything, guys. I didn't even alt-tab. Codex just opened it. It It just opened my browser. It's like, "Yo, just look at this, okay? These agents are getting scary...”
The MP3 waveform visualizer shows the real leverage: once built, a composition is a reusable asset you tweak in minutes instead of rebuilding, and Codex's pre-send feature lets you queue feedback (like 'the waveform hits the top too soon') that auto-sends when the current run finishes — the lasting advantage is taste and judgment about which animations are good, not raw production. Build one reusable composition (a waveform, lower third, or logo animation), then pre-send one refinement prompt while it renders and save the result to your own asset library.
01
Brief
Start with this video's job: This video shows how to set up the Hyperframes plugin inside OpenAI's Codex app so an AI agent can produce and edit videos by writing HTML code — the timeline itself becomes code — covering the new HTML-in-canvas feature, parallel threads via built-in git worktrees, and real builds like a liquid-glass subscribe animation, a website product demo, and an MP3 waveform visualizer. Treat "Brief" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:58, where the video says: “Premiere, Final Cut, CapCut, these all require a human dragging clips on a timeline. Now, Hyperframes completely flips this because the timeline itself becomes code. And now, an AI agent like Codex, Hermes, Claude, they can produce, edit,...”
02
Source
Use "Source" to locate the part of the creative automation workflow the video is demonstrating. Ask what changes in your real setup if this claim is true. Anchor it to 11:04, where the video says: “sure to use {slash} compact a lot, okay? This is a built-in command into Codex that will compact the chat history to save you tokens because this can eat up your limits very fast. Now, obviously, Codex has...”
03
Generation
Turn "Generation" into the reusable artifact for this lesson: A creative workflow board with critique criteria and review checkpoints. This is where watching becomes something you can inspect and reuse.
04
Selection
Use "Selection" 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
Edit
Use "Edit" 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
Taste Review
Use "Taste Review" 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 creative workflow board with critique criteria and review checkpoints..
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 how to set up the Hyperframes plugin inside OpenAI's Codex app so an AI agent can produce and edit videos by writing HTML code — the timeline itself becomes code — covering the new HTML-in-canvas feature, parallel threads via built-in git worktrees, and real builds like a liquid-glass subscribe animation, a website product demo, and an MP3 waveform visualizer.
02
Explain the practical stakes without hype: New playlist item from David Ondrej; queued for transcript-backed review, topic mapping, and a practical learning artifact.
03
Map the idea onto the Brief -> Source -> Generation -> Selection -> Edit -> Taste Review sequence and name the weakest link.
04
Produce the artifact and include the evidence that proves it: A creative workflow board with critique criteria and review checkpoints.
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: Codex can now make videos… it’s insane
- URL: https://www.youtube.com/watch?v=oyWSdPYeQwQ
- Topic: Creative Automation
- My current learning frame: Set up Codex with the Hyperframes plugin, enable the canvas-draw-element Chrome flag, and one-shot three assets — an explainer animation, a subscribe button, and a product demo from your own website URL — using parallel threads and /compact along the way.
- Why this matters: New playlist item from David Ondrej; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:58 / Evidence 1: "Premiere, Final Cut, CapCut, these all require a human dragging clips on a timeline. Now, Hyperframes completely flips this because the timeline itself becomes code. And now, an AI agent like Codex, Hermes, Claude, they can produce, edit,..."
- 5:08 / Evidence 2: "Chrome or Brave, open a new tab, paste this in, and you need to have this enabled. As you can see, by default, it's disabled. So, let's enable it. And then, you need to relaunch the browser. Now,..."
- 8:38 / Evidence 3: "want to make changes, right? Well, that's as easy as prompting Codex. So, let's jump back in and say, "Okay, but change the design to be kind of a red gradient vibe. Do not change anything else." And..."
- 11:04 / Evidence 4: "sure to use {slash} compact a lot, okay? This is a built-in command into Codex that will compact the chat history to save you tokens because this can eat up your limits very fast. Now, obviously, Codex has..."
- 16:18 / Evidence 5: "see, Codex just opens this. I didn't even do anything, guys. I didn't even alt-tab. Codex just opened it. It It just opened my browser. It's like, "Yo, just look at this, okay? These agents are getting scary..."
- 17:49 / Evidence 6: "you can see in in the Codex app, you can just work on multiple projects in parallel, switching no problem. Uh let's see how this one is doing. Okay, almost 400 lines of code. Yeah, this is still..."
- 20:43 / Evidence 7: "different rather than you know, hitting the top so quickly and so easily. And I sent it and it's going to Once it finishes the previous it's going to auto-send the next prompt. So, when you get an..."
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 creative workflow board with critique criteria and review checkpoints.
5. Include:
- a plain-English definition of the core idea
- a diagram or structured model using this sequence: Brief -> Source -> Generation -> Selection -> Edit -> Taste Review
- 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 "Codex can now make videos… it’s insane", not a generic Creative Automation 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.
Creative AI removes the need for taste.
It increases the need for taste because output volume explodes.
The best prompt is enough.
References, critique, iteration, and post-production matter just as much.
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 creative workflow board with critique criteria and review checkpoints..
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.
Why does Hyperframes let AI agents edit video when traditional tools like Premiere or CapCut could not?
What Codex feature lets you run a second Hyperframes prompt while the first is still generating, without interference?
What is Codex's pre-send feature and why is it useful during long Hyperframes runs?
Source shelf
Use the video as a doorway, then verify with primary sources.
