StarZero All shows Try the podcast clipper →

Jun 30, 2025 · 51m · startup-ideas

This AI Agent creates 1000+ SEO Pages in 52 min (Claude + MCP + Cursor)

Greg Isenberg · 6m spoken
0:00 / 0:00
▶ Watch on YouTube →

gold bands on the timeline = statements, start to end. Hover to read, click to jump. CC turns on captions

Greg Isenberg hosts James (The Boring Marketer) for a practical, step-by-step masterclass on using Cursor, Claude Code, and Model Context Protocol (MCP) integrations to research, build, and deploy high-traffic programmatic SEO web applications without traditional engineering skills.

How this conversation actually went

Every chapter scored 0–10 on four independent dynamics. Hover any point for the reasoning behind the score. Greg holds 13.5% of the talking time here. How this is scored →

Greg as informed peer 3.1 Guest teaching 5.3 Guest disagreement 0.1 Greg pushing back 0.0
05100:0015:0030:0045:001:29–6:12 · Greg as informed peer 0/10 Navigating the Cursor Workspace and Integrating Claude Code James provides a detailed walkthrough of his Cursor and Claude Code workspace setup, explaining why he utilizes the terminal interface over native agent features. Greg does not participate in this opening demonstration segment.6:12–10:34 · Greg as informed peer 1/10 Sponsor Segment: Managing Business Finances with Brex Following Greg's sponsor read, James explains how simple it is to deploy projects via Vercel and set up MCPs using natural language prompts. Greg offers encouraging enthusiasm without challenging any claims.10:35–21:53 · Greg as informed peer 4/10 Automated Keyword Discovery Using Firecrawl and Perplexity MCPs James demonstrates using Firecrawl and Perplexity MCPs to discover programmatic SEO opportunities and scrape domain context. Greg adds relevant commentary regarding typical agency costs and the necessity of conversational back-and-forth prompting.21:53–25:04 · Greg as informed peer 5/10 Formulating a Product Requirement Document for Programmatic Pages Greg asks why a PRD is generated at this exact phase, contributing his own insight that lacking a PRD dilutes AI output whereas drip-feeding context yields higher quality results. James validates this and integrates the document into Claude Code.25:05–29:21 · Greg as informed peer 3/10 Orchestrating the Web Application Build with Claude Code Claude Code begins orchestrating the web application build based on the structured PRD to-do list. Greg pauses the live demo to reflect on how non-technical founders can literally program full applications using prompt-driven workflows.29:22–33:57 · Greg as informed peer 3/10 The Advantage of AI Coding Over Traditional Workflows Greg raises the prevalent trend of n8n automations on social media, prompting James to critically contrast brittle node workflows with Claude Code micro-apps. James dismisses viral workflow giveaways as frequently non-functional.33:58–42:49 · Greg as informed peer 5/10 Managing Development Costs and Bridging Figma with Tailwind James shows how he converted Figma designs into Tailwind code via Anima and checks his token expenditure with slash cost. Greg delivers a strong counter-perspective regarding the massive ROI of micro-apps versus traditional university coding education.42:50–49:29 · Greg as informed peer 4/10 Deploying to Vercel and Reviewing Programmatic Pages James deploys the finished programmatic comparison hub to Vercel and reviews its dynamically generated SEO features and lead magnet. Greg extends the discussion by pointing out the monetization potential of affiliate links on comparison pages.1:29–6:12 · Guest teaching 5/10 Navigating the Cursor Workspace and Integrating Claude Code James provides a detailed walkthrough of his Cursor and Claude Code workspace setup, explaining why he utilizes the terminal interface over native agent features. Greg does not participate in this opening demonstration segment.6:12–10:34 · Guest teaching 4/10 Sponsor Segment: Managing Business Finances with Brex Following Greg's sponsor read, James explains how simple it is to deploy projects via Vercel and set up MCPs using natural language prompts. Greg offers encouraging enthusiasm without challenging any claims.10:35–21:53 · Guest teaching 6/10 Automated Keyword Discovery Using Firecrawl and Perplexity MCPs James demonstrates using Firecrawl and Perplexity MCPs to discover programmatic SEO opportunities and scrape domain context. Greg adds relevant commentary regarding typical agency costs and the necessity of conversational back-and-forth prompting.21:53–25:04 · Guest teaching 5/10 Formulating a Product Requirement Document for Programmatic Pages Greg asks why a PRD is generated at this exact phase, contributing his own insight that lacking a PRD dilutes AI output whereas drip-feeding context yields higher quality results. James validates this and integrates the document into Claude Code.25:05–29:21 · Guest teaching 5/10 Orchestrating the Web Application Build with Claude Code Claude Code begins orchestrating the web application build based on the structured PRD to-do list. Greg pauses the live demo to reflect on how non-technical founders can literally program full applications using prompt-driven workflows.29:22–33:57 · Guest teaching 6/10 The Advantage of AI Coding Over Traditional Workflows Greg raises the prevalent trend of n8n automations on social media, prompting James to critically contrast brittle node workflows with Claude Code micro-apps. James dismisses viral workflow giveaways as frequently non-functional.33:58–42:49 · Guest teaching 5/10 Managing Development Costs and Bridging Figma with Tailwind James shows how he converted Figma designs into Tailwind code via Anima and checks his token expenditure with slash cost. Greg delivers a strong counter-perspective regarding the massive ROI of micro-apps versus traditional university coding education.42:50–49:29 · Guest teaching 6/10 Deploying to Vercel and Reviewing Programmatic Pages James deploys the finished programmatic comparison hub to Vercel and reviews its dynamically generated SEO features and lead magnet. Greg extends the discussion by pointing out the monetization potential of affiliate links on comparison pages.1:29–6:12 · Guest disagreement 0/10 Navigating the Cursor Workspace and Integrating Claude Code James provides a detailed walkthrough of his Cursor and Claude Code workspace setup, explaining why he utilizes the terminal interface over native agent features. Greg does not participate in this opening demonstration segment.6:12–10:34 · Guest disagreement 0/10 Sponsor Segment: Managing Business Finances with Brex Following Greg's sponsor read, James explains how simple it is to deploy projects via Vercel and set up MCPs using natural language prompts. Greg offers encouraging enthusiasm without challenging any claims.10:35–21:53 · Guest disagreement 0/10 Automated Keyword Discovery Using Firecrawl and Perplexity MCPs James demonstrates using Firecrawl and Perplexity MCPs to discover programmatic SEO opportunities and scrape domain context. Greg adds relevant commentary regarding typical agency costs and the necessity of conversational back-and-forth prompting.21:53–25:04 · Guest disagreement 0/10 Formulating a Product Requirement Document for Programmatic Pages Greg asks why a PRD is generated at this exact phase, contributing his own insight that lacking a PRD dilutes AI output whereas drip-feeding context yields higher quality results. James validates this and integrates the document into Claude Code.25:05–29:21 · Guest disagreement 0/10 Orchestrating the Web Application Build with Claude Code Claude Code begins orchestrating the web application build based on the structured PRD to-do list. Greg pauses the live demo to reflect on how non-technical founders can literally program full applications using prompt-driven workflows.29:22–33:57 · Guest disagreement 1/10 The Advantage of AI Coding Over Traditional Workflows Greg raises the prevalent trend of n8n automations on social media, prompting James to critically contrast brittle node workflows with Claude Code micro-apps. James dismisses viral workflow giveaways as frequently non-functional.33:58–42:49 · Guest disagreement 0/10 Managing Development Costs and Bridging Figma with Tailwind James shows how he converted Figma designs into Tailwind code via Anima and checks his token expenditure with slash cost. Greg delivers a strong counter-perspective regarding the massive ROI of micro-apps versus traditional university coding education.42:50–49:29 · Guest disagreement 0/10 Deploying to Vercel and Reviewing Programmatic Pages James deploys the finished programmatic comparison hub to Vercel and reviews its dynamically generated SEO features and lead magnet. Greg extends the discussion by pointing out the monetization potential of affiliate links on comparison pages.1:29–6:12 · Greg pushing back 0/10 Navigating the Cursor Workspace and Integrating Claude Code James provides a detailed walkthrough of his Cursor and Claude Code workspace setup, explaining why he utilizes the terminal interface over native agent features. Greg does not participate in this opening demonstration segment.6:12–10:34 · Greg pushing back 0/10 Sponsor Segment: Managing Business Finances with Brex Following Greg's sponsor read, James explains how simple it is to deploy projects via Vercel and set up MCPs using natural language prompts. Greg offers encouraging enthusiasm without challenging any claims.10:35–21:53 · Greg pushing back 0/10 Automated Keyword Discovery Using Firecrawl and Perplexity MCPs James demonstrates using Firecrawl and Perplexity MCPs to discover programmatic SEO opportunities and scrape domain context. Greg adds relevant commentary regarding typical agency costs and the necessity of conversational back-and-forth prompting.21:53–25:04 · Greg pushing back 0/10 Formulating a Product Requirement Document for Programmatic Pages Greg asks why a PRD is generated at this exact phase, contributing his own insight that lacking a PRD dilutes AI output whereas drip-feeding context yields higher quality results. James validates this and integrates the document into Claude Code.25:05–29:21 · Greg pushing back 0/10 Orchestrating the Web Application Build with Claude Code Claude Code begins orchestrating the web application build based on the structured PRD to-do list. Greg pauses the live demo to reflect on how non-technical founders can literally program full applications using prompt-driven workflows.29:22–33:57 · Greg pushing back 0/10 The Advantage of AI Coding Over Traditional Workflows Greg raises the prevalent trend of n8n automations on social media, prompting James to critically contrast brittle node workflows with Claude Code micro-apps. James dismisses viral workflow giveaways as frequently non-functional.33:58–42:49 · Greg pushing back 0/10 Managing Development Costs and Bridging Figma with Tailwind James shows how he converted Figma designs into Tailwind code via Anima and checks his token expenditure with slash cost. Greg delivers a strong counter-perspective regarding the massive ROI of micro-apps versus traditional university coding education.42:50–49:29 · Greg pushing back 0/10 Deploying to Vercel and Reviewing Programmatic Pages James deploys the finished programmatic comparison hub to Vercel and reviews its dynamically generated SEO features and lead magnet. Greg extends the discussion by pointing out the monetization potential of affiliate links on comparison pages.

speaking balance: gold is Greg, purple is the guest (3 minute bins)

0:00 · Greg 40.6% · guest 59.4%0:00 · Greg 40.6% · guest 59.4%3:00 · Greg 0% · guest 100%3:00 · Greg 0% · guest 100%6:00 · Greg 40.8% · guest 59.2%6:00 · Greg 40.8% · guest 59.2%9:00 · Greg 2.3% · guest 97.7%9:00 · Greg 2.3% · guest 97.7%12:00 · Greg 0.4% · guest 99.6%12:00 · Greg 0.4% · guest 99.6%15:00 · Greg 16.2% · guest 83.8%15:00 · Greg 16.2% · guest 83.8%18:00 · Greg 2.4% · guest 97.6%18:00 · Greg 2.4% · guest 97.6%21:00 · Greg 16.3% · guest 83.7%21:00 · Greg 16.3% · guest 83.7%24:00 · Greg 0% · guest 100%24:00 · Greg 0% · guest 100%27:00 · Greg 5.3% · guest 94.7%27:00 · Greg 5.3% · guest 94.7%30:00 · Greg 10.4% · guest 89.6%30:00 · Greg 10.4% · guest 89.6%33:00 · Greg 13.8% · guest 86.2%33:00 · Greg 13.8% · guest 86.2%36:00 · Greg 5% · guest 95%36:00 · Greg 5% · guest 95%39:00 · Greg 31% · guest 69%39:00 · Greg 31% · guest 69%42:00 · Greg 8.4% · guest 91.6%42:00 · Greg 8.4% · guest 91.6%45:00 · Greg 1.8% · guest 98.2%45:00 · Greg 1.8% · guest 98.2%48:00 · Greg 14.7% · guest 85.3%48:00 · Greg 14.7% · guest 85.3%51:00 · Greg 78.5% · guest 21.5%51:00 · Greg 78.5% · guest 21.5%
Sharpest disagreement ▶ 32:25 Calling out fake social media n8n giveaways

James breaks from his collaborative tone to call out the dirty secret behind viral social media n8n automation giveaways, noting that creators often upload non-functional LLM outputs.

 Hardest push from Greg ▶ 33:58 Pushing back on the intimidation barrier of terminal coding

Greg interjects to balance James's enthusiasm, highlighting the necessary caveat that jumping into code and terminal environments remains deeply jarring for non-technical individuals.

 Biggest teaching moment ▶ 37:54 Bridging Figma designs to Tailwind with Anima

James walks through an actionable workflow using the Anima plugin to extract Tailwind CSS layer-by-layer from Figma, enabling Claude Code to replicate exact brand designs.

 Greg holds their own ▶ 41:38 Defending AI token costs against traditional education

Greg proactively addresses anticipated audience skepticism regarding API expenses by comparing twenty-dollar app builds against hundred-thousand-dollar computer science degrees and business ROI.
the scores for every segment, with the reasoning behind each
ChapterTopicGreg as informed peerGuest teachingGuest disagreementGreg pushing backWhy
Navigating the Cursor Workspace and Integrating Claude Code 0500 James provides a detailed walkthrough of his Cursor and Claude Code workspace setup, explaining why he utilizes the terminal interface over native agent features. Greg does not participate in this opening demonstration segment.
Sponsor Segment: Managing Business Finances with Brex 1400 Following Greg's sponsor read, James explains how simple it is to deploy projects via Vercel and set up MCPs using natural language prompts. Greg offers encouraging enthusiasm without challenging any claims.
Automated Keyword Discovery Using Firecrawl and Perplexity MCPs 4600 James demonstrates using Firecrawl and Perplexity MCPs to discover programmatic SEO opportunities and scrape domain context. Greg adds relevant commentary regarding typical agency costs and the necessity of conversational back-and-forth prompting.
Formulating a Product Requirement Document for Programmatic Pages 5500 Greg asks why a PRD is generated at this exact phase, contributing his own insight that lacking a PRD dilutes AI output whereas drip-feeding context yields higher quality results. James validates this and integrates the document into Claude Code.
Orchestrating the Web Application Build with Claude Code 3500 Claude Code begins orchestrating the web application build based on the structured PRD to-do list. Greg pauses the live demo to reflect on how non-technical founders can literally program full applications using prompt-driven workflows.
The Advantage of AI Coding Over Traditional Workflows 3610 Greg raises the prevalent trend of n8n automations on social media, prompting James to critically contrast brittle node workflows with Claude Code micro-apps. James dismisses viral workflow giveaways as frequently non-functional.
Managing Development Costs and Bridging Figma with Tailwind 5500 James shows how he converted Figma designs into Tailwind code via Anima and checks his token expenditure with slash cost. Greg delivers a strong counter-perspective regarding the massive ROI of micro-apps versus traditional university coding education.
Deploying to Vercel and Reviewing Programmatic Pages 4600 James deploys the finished programmatic comparison hub to Vercel and reviews its dynamically generated SEO features and lead magnet. Greg extends the discussion by pointing out the monetization potential of affiliate links on comparison pages.

Statements from this episode (1)

Insight
Greg Isenberg: Structuring AI prompts with a PRD prevents diluted outputs
“I think if you, what I've noticed is if you don't create a PRD, the output ends up being worse, like very diluted. So what's cool about the PRD is you can kind of like drip feed context and because of that, because of how tight the drip feeding is, you end up …”
Greg Isenberg Jun 30, 2025 ▶ 23:28
Made with StarZero

Turn any episode into a week of clips.

This entire site, over 300 episodes transcribed, diarized, checked and made playable, runs on the StarZero media pipeline. Drop in your own episode and the podcast clipper finds the moments worth sharing, cuts them, captions them, and reframes them for every feed.

Try the podcast clipper Learn more