A hands-on demo of pairing OpenAI's Codex with Paper (a Figma-like canvas exposed over an MCP server) to produce real design work in natural language—a full Avis brand refresh, LinkedIn carousels, a Tanner Goods marketing email, a SaaS website redesign, and a mobile habit-tracker—using a strategy-doc-then-design two-prompt workflow and GPT image gen for mockups.
Pat SimmonsWatchTranscript 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 Pat Simmons; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Skill you build: The ability to drive a design agent (Codex + Paper over MCP) through a strategy-first, two-prompt workflow—and to keep prodding it past its lazy first pass—to ship multiple distinct, on-brand design directions instead of one generic template.
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.
5,074 cleaned transcript words reviewed across 1,447 timed caption segments.
Thesis
Codex + Paper = INCREDIBLE Designs teaches a practical codex + claude workflows move: A hands-on demo of pairing OpenAI's Codex with Paper (a Figma-like canvas exposed over an MCP server) to produce real design work in natural language—a full Avis brand refresh, LinkedIn carousels, a Tanner Goods marketing email, a SaaS website redesign, and a mobile habit-tracker—using a strategy-doc-then-design two-prompt workflow and GPT image gen for mockups.
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.
0:13
Wire up Codex + Paper
“entire mobile app experiences. And the beauty of all of this, guess what? It can be done by just talking to Codex in natural language. So let me show you how you too can become a professional designer...”
The setup is a separation-of-concerns workflow: one Codex agent writes a strategy doc first, then a second agent uses that doc to build in Paper via its MCP server. Connecting it means downloading the Paper desktop app (the browser version won't drive the MCP), adding Paper as a Streamable HTTP MCP server in Codex settings with no bearer token, and restarting Codex—then a 'create a red box' test confirms the canvas link is live. Download Paper, add its Streamable HTTP MCP endpoint to your coding agent, restart, and verify the connection by asking the agent to create a single artboard or red box on the canvas before attempting any real design.
12:15
Prod the lazy agent
“free of charge. So, I wrote this prompt. I'm going to go into Codex. We're going to go to new chat. And again, just to save our designer context, we're first going to write the copy. So, I...”
The agent does the least work it can and calls a job done, so quality comes from pushing back: he repeatedly tells it 'work harder,' 'what's missing,' and 'are these three directions truly distinct,' then has it QA its own output. He also writes copy first in a separate prompt to save the design agent's context, and uses GPT image gen 2 for photorealistic mockups that replace weeks of Photoshop work. Take any first-pass design the agent produces and run three rounds of specific critique—name what's missing, demand the variations be genuinely distinct, then have it QA itself—and note how much the output improves with each prod.
15:35
Brand-absorb then ship
“website, whatever you need to do to like give this creative brief to to a designer." Okay, we've got this prompt to give to our design agent. Go here, paste this. In case you're curious, this is the...”
For a Tanner Goods email he points the agent at the live homepage and products to absorb the real visual identity and product names, generates fresh product imagery with GPT image gen, and asks for three deliberately ranged directions (one restrained, one bolder, one unexpected). Paper lets him hand-fix details like a line break directly on the canvas, and the result can be exported as HTML and pushed to a sender like Resend—skipping Mailchimp-style templates entirely. Pick a real brand's live site, prompt your agent to absorb its visual identity and pull actual product names, then request three ranged email directions (restrained, bold, unexpected) and export your favorite as HTML.
01
Inspect
Start with this video's job: A hands-on demo of pairing OpenAI's Codex with Paper (a Figma-like canvas exposed over an MCP server) to produce real design work in natural language—a full Avis brand refresh, LinkedIn carousels, a Tanner Goods marketing email, a SaaS website redesign, and a mobile habit-tracker—using a strategy-doc-then-design two-prompt workflow and GPT image gen for mockups. Treat "Inspect" as the outcome you are trying to make visible, not a topic label. Anchor it to 0:13, where the video says: “entire mobile app experiences. And the beauty of all of this, guess what? It can be done by just talking to Codex in natural language. So let me show you how you too can become a professional designer...”
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 12:15, where the video says: “free of charge. So, I wrote this prompt. I'm going to go into Codex. We're going to go to new chat. And again, just to save our designer context, we're first going to write the copy. So, I...”
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: A hands-on demo of pairing OpenAI's Codex with Paper (a Figma-like canvas exposed over an MCP server) to produce real design work in natural language—a full Avis brand refresh, LinkedIn carousels, a Tanner Goods marketing email, a SaaS website redesign, and a mobile habit-tracker—using a strategy-doc-then-design two-prompt workflow and GPT image gen for mockups.
02
Explain the practical stakes without hype: New playlist item from Pat Simmons; 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: Codex + Paper = INCREDIBLE Designs
- URL: https://www.youtube.com/watch?v=a8xPIfJpEOI
- Topic: Codex + Claude Workflows
- My current learning frame: Pick one recurring design chore (an email, a carousel, or a landing page) for a real brand, run the two-prompt Codex+Paper workflow—strategy doc first, then build via the Paper MCP—prod the agent through several critique rounds for three distinct on-brand directions, and export the best one as HTML.
- Why this matters: New playlist item from Pat Simmons; queued for transcript-backed review, topic mapping, and a practical learning artifact.
Transcript anchors from this exact video:
- 0:13 / Evidence 1: "entire mobile app experiences. And the beauty of all of this, guess what? It can be done by just talking to Codex in natural language. So let me show you how you too can become a professional designer..."
- 2:03 / Evidence 2: "I'll show you exactly what I searched. paper.design MCP server. And then you click on the docs page. They explain what an MCP server is and how to connect to each of your different platforms, Codex, Antigravity, Open..."
- 3:50 / Evidence 3: "into a brand refresh. Maybe the strategy doc says what should be included, but I'm just going to have agents put all these pieces together. We're going to go medium cuz I'm going to max out my Codex..."
- 8:33 / Evidence 4: "going to immediately stand out on LinkedIn, which as you know is not that hard to do. So, go over to Codex, go to new chat, and I'll paste in this prompt. So, the prompt is I'm making..."
- 12:15 / Evidence 5: "free of charge. So, I wrote this prompt. I'm going to go into Codex. We're going to go to new chat. And again, just to save our designer context, we're first going to write the copy. So, I..."
- 15:35 / Evidence 6: "website, whatever you need to do to like give this creative brief to to a designer." Okay, we've got this prompt to give to our design agent. Go here, paste this. In case you're curious, this is the..."
- 22:06 / Evidence 7: "there's one layer that Paper doesn't handle, motion. It has some of these motion built in here, like these, for example, but it doesn't have a whole lot. So, watch this video next where I run through Codex..."
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 "Codex + Paper = INCREDIBLE Designs", 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.
To connect Paper to Codex via MCP, what specific app must you install and why won't the alternative work, and which MCP server type/auth do you select?
The presenter says the agent is lazy and does the least work possible. What specific prods does he repeatedly use to raise quality, and what does he have the agent do at the end?
For the Tanner Goods email, how does he get the agent to capture the real brand, and how does he frame the three design directions he asks for?
Source shelf
Use the video as a doorway, then verify with primary sources.
