Aakash GuptaUse these skills to supercharge your claude code setup | Oji Udezue | 3x CPO
At a glance
WHAT IT’S REALLY ABOUT
Product judgment skills in Claude Code for builder PM workflows
- Oji Udezue argues PMs are becoming the bottleneck as engineering speed increases, and proposes “multi-skill” Claude Code workflows that compress business, product, and engineering steps into one repo.
- He demos a New Project Scaffolding skill that runs an 11-step process: viability gate, market research, product brief/PRD artifacts, architecture decisions, repo structure, testing, CI/CD, and security hygiene.
- The viability gate is positioned as a rare LLM behavior that can explicitly recommend “don’t build this,” using structured PM frameworks rather than generic encouragement.
- A second demo idea (“Standup Zero”) scores weak/moderate due to shallow urgency, unclear target user, heavy competition, and weak differentiation—illustrating how the system saves time by killing ideas early.
- After an idea passes, the workflow shifts to product layer execution—interaction model questions, rapid UI prototypes, and a customer discovery plan that includes scripts, survey, synthesis templates, and recruiting gates.
IDEAS WORTH REMEMBERING
5 ideasSpeeding up engineering forces PMs to upgrade their “judgment throughput.”
Udezue describes developers accelerating with AI while PM discovery and decision-making lags, making PMs the constraint unless they adopt structured, repeatable LLM-assisted workflows.
Treat product, business, and code as a single workflow—not separate tools and teams.
The scaffolding approach collapses typical separation (Notion PRDs, Figma, dev repos) into one GitHub repo that contains artifacts, context, and production-ready engineering scaffolds.
A viability gate is valuable because it can say “no” early and explicitly.
The skill scores dimensions like problem clarity/urgency, target user, competition, differentiation, feasibility, and revenue potential, and will recommend stopping when weaknesses stack up—countering LLM optimism bias.
LLM market research can be 60–70% helpful, but must be audited for credibility.
They recommend a “PRD review” mindset: check whether competitors and trends are current, whether the differentiation map makes sense, and whether the brief matches what you know about the space.
A strong scaffold includes quality systems, not just starter code.
Beyond generating docs, the skill creates CLAUDE.md tailored to the project, a self-organizing folder structure, tests, CI templates, security/secret-handling guidance, and bug/learning practices—common gaps for vibe coders.
WORDS WORTH SAVING
5 quotesThe era of mono-skilled professionals is dead, so I'm going to demonstrate the future of product builders.
— Oji Udezue
I've been a product manager for 25 years, and I've never really felt that coding was worth my time until now.
— Oji Udezue
LLMs very rarely tell you no, and this skill will tell you absolutely no, don't do this.
— Oji Udezue
People build, and then there's no customer for it. There's nobody for it. In fact, it's super puzzling right now because it feels like GitHub is the only place that people can express themselves when they have Claude Code. They just build anything, put it on GitHub, and just languish with zero stars, if that makes any sense.
— Oji Udezue
The plan itself is complete and executable. Every script, the survey, the synthesis template, and the weeks time budgeted below. What is missing is the one input the skill will not let a plan proceed. A confirmed list of recruitable people with real target cohorts, um, not adjacent contacts from the founder's network.
— Oji Udezue
High quality AI-generated summary created from speaker-labeled transcript.
