Agent skill

Product Taste

by RefoundAI in RefoundAI/lenny-skills

Help users build a professional eye for quality, articulate the reasoning behind successful product paradigms, and develop the conviction to make high-stakes decisions before data is available.

MITAuto-check passed

Install Product Taste

skills CLI
$ npx skills add RefoundAI/lenny-skills --skill product-taste -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install RefoundAI/lenny-skills product-taste --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/RefoundAI/lenny-skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/product-taste .claude/skills/product-taste && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
product-taste
GitHub stars
1.4k
Token cost
~2k tokens
SKILL.md length
1,208 words
Files
3 (incl. references)
Skills in repo
76
Repo updated
First seen
Licence
MIT

At a glance

Help users build a professional eye for quality, articulate the reasoning behind successful product paradigms, and develop the conviction to make high-stakes decisions before data is available.

  • Works in 4 steps: Deconstruct experiences - Guide the user… → Calibrate intuition - Help the user… → Challenge assumptions - Use… → …
  • SKILL.md covers How to Help, Core Principles, Templates & Frameworks and Questions to Help Users, plus 3 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Product Taste is an agent skill from RefoundAI/lenny-skills. Help users build a professional eye for quality, articulate the reasoning behind successful product paradigms, and develop the conviction to make high-stakes decisions before data is available.

Its SKILL.md is about 2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 3 other files, including reference files (for example `references/artifacts.md` and `references/guest-insights.md`).

The repository describes itself as: 86 product management skills from Lenny's Podcast for Claude Code and AI agents. Hiring, user research, strategy, shipping, and more. The licence is MIT.

Example prompts

  • “/product-taste”

Workflow steps

4 steps, taken from the first numbered list in SKILL.md.

  1. Deconstruct experiences - Guide the user through analyzing specific product details to identify why they evoke certain emotional responses.
  2. Calibrate intuition - Help the user document and review their decision rationales against real-world outcomes to refine their judgment.
  3. Challenge assumptions - Use first-principles thinking to question standard industry metrics and focus on core user utility.
  4. Refine the feel - Assist in articulating the specific polish, speed, and usability elements that contribute to a high-quality product feel.

What it can do on your machine

Read from SKILL.md and the folder at commit 13598cc. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Product Taste loads about 2k tokens when it runs, and up to ~26k if it reads all its reference files. Until then it costs about 52 tokens; SKILL.md has 1,208 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~52
When it runs · the whole SKILL.md, loaded when a task matches
~2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~26k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from RefoundAI/lenny-skills at commit 13598cc, republished under its MIT licence (© RefoundAI). 1,208 words, ~2,017 tokens.

Download SKILL.mdSave it as .claude/skills/product-taste/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
product-taste
description
Help users build a professional eye for quality, articulate the reasoning behind successful product paradigms, and develop the conviction to make high-stakes decisions before data is available.

Product Taste and Intuition

Develop a reliable internal compass to recognize and build world-class products.

Help the user with product taste and intuition using insights from 35 guests and posts across Lenny's Podcast and Newsletter.

How to Help

  1. Deconstruct experiences - Guide the user through analyzing specific product details to identify why they evoke certain emotional responses.
  2. Calibrate intuition - Help the user document and review their decision rationales against real-world outcomes to refine their judgment.
  3. Challenge assumptions - Use first-principles thinking to question standard industry metrics and focus on core user utility.
  4. Refine the feel - Assist in articulating the specific polish, speed, and usability elements that contribute to a high-quality product feel.

Core Principles

Learn Through Model Primitives

Howie Liu: "I think to really understand the solution space of what's possible, you have to be in the details. I mean, literally, you can't just look at screenshots or a pre-recorded video of a new product feature. AI is something you have to play with, and ideally you're playing with both the packaged up app or solution that you've built with it, but you're also playing around directly with the underlying primitives who are using the models either via API or via a chat interface."

True intuition is built by playing directly with the underlying technology and APIs rather than observing demos. Understanding the raw boundaries of your ingredients allows you to imagine new possibilities.

Articulate Emotional Responses

Jessica Hische: "Most people are better at understanding the feelings and sensations that typography and logos give us than they give themselves credit for, because what we are as people are endless absorbers of patterns, and information, and all this kind of stuff as we move throughout the world. We don't take time to sit and digest it, but it's still coming in and getting logged, and so even as a non-designer, I think you can look at examples of logos where something's not quite right and be like, 'Something's not right here, I just don't know how to name it.'"

Develop a professional eye by explicitly naming the feelings triggered by specific design elements. Practice identifying broken patterns to understand what makes a visual work.

Design for Human Emotion

Josh Miller: "What we do at The Browser Company is we talk about optimizing feelings. How do we want to make someone feel on the other end of our software? Do we want to make them feel joy? Do we want to make them feel fast?"

Focus development on evoking specific human feelings like joy or speed rather than just moving quantitative metrics. Treat metrics as secondary tools to measure the success of an emotional experience.

Actively Falsify Assumptions

Judd Antin: "One of my big mantras was, "We don't validate, we falsify. We are looking to be wrong." Many PMs, many designers are not in that place. They do not want to be wrong. They're looking to validate, and that's user-centered performance."

Use research to stress-test your internal beliefs and look for ways you might be wrong. Avoid research that merely validates existing decisions without the potential to change them.

Quality as a Growth Strategy

Katie Dill: "I know there's this saying of it's growth versus quality, but quality is growth. And if you think about how you can make your product easier to use and more understandable, that will of course drive people to use it, and use more of it, and have a better experience with it that they'll want to talk about with others."

Superior aesthetics and usability are not just for show; they drive user activation and word-of-mouth. High-quality details make a product feel more approachable and trustworthy.

Iterate via Internal Usage

Karri Saarinen: "We actually believe that when you start building the thing you actually start realizing more how it should work and how it should be better. A lot of times with the teams we tell them, "Just put it there in, I don't know, the first week almost. After you have some designs in place or some design ideas, just put it into the app and ship it to production." It's only visible to us so we internally can test it out."

Ship early to internal teams and co-create with select customers to find real usage patterns. High quality is reached through constant iteration before final polishing.

Show full SKILL.md (485 more words)Show less

Templates & Frameworks

  • Three Dimensions of Product Feel (What working at Figma taught me about customer obsession) - The three specific dimensions Figma focused on to get the 'feel' of their product right, applicable to any high-frequency software product
  • Signs Your Product Sense Is Improving (How to develop product sense) - Six indicators that your product sense is developing over time, useful as a self-assessment rubric.
  • Competitive Product Comparison Template (How to develop product sense) - A structured comparison framework for analyzing two competing products across multiple dimensions to uncover strategic differences and design choices.
  • Shopify's Product Quality Benchmarking Questions (How Shopify builds product) - Questions to ask when evaluating whether your product meets a high quality bar
  • Mona Lisa Principle (How Miro builds product) - A quality standard principle stating that everything shipped should be something the team would be proud to put their name on, like the Mona Lisa painting
  • Adapt When Adopting Framework (How Duolingo reignited user growth) - A three-question framework for evaluating whether to borrow a feature from another product and how to adapt it to your own context
  • Three Key Features Framework (Great ≠ Good) (Essential reading for product builders—part 1) - Paul Buchheit's product design principle: pick three key attributes or features for a new product, get those very right, and forget everything else. If a produc
  • Onion-Peeling Product Philosophy (How Notion builds product) - A metaphor for building products that serve both casual users and power users by letting them go as deep into customization as they want

See references/artifacts.md for the full list with details.

Questions to Help Users

  • "What is the specific emotion you want this feature to evoke in the user?"
  • "If you couldn't use any data or metrics, what would your gut tell you to change right now?"
  • "How does the feel of this product compare to the most delightful tools you use daily?"
  • "What are the core technical primitives of this feature, and have you interacted with them directly?"
  • "Can you identify any broken or friction-filled patterns in the current user flow?"
  • "Is this feature solving a root problem or just responding to a surface-level request?"

Common Mistakes to Flag

  • Metric-Obsessed Tunnel Vision - Relying solely on quantitative data can lead to a soulless product that lacks a cohesive, opinionated vision.
  • Blind Deconstruction - Copying successful features from competitors without understanding their strategic context often results in poor fit for your own users.
  • Seeking Only Validation - Using research only to confirm what you already believe prevents you from finding the fatal flaws in your strategy.
  • The Quality-Skill Gap Stall - Waiting for your skills to match your high standards instead of producing a high volume of work to close that gap through practice.
  • Over-Optimizing Hidden Metrics - Focusing on backend metrics that users can't see while neglecting the visible innovation that signals brand growth.

Deep Dive

For all 36 sourced insights from 35 guests, see references/guest-insights.md

  • Customer Interviews
  • Continuous Discovery
  • Idea Validation
  • Product Experiments

© RefoundAI, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 2 other files (references) in skills/product-taste of RefoundAI/lenny-skills.

  • SKILL.md
  • references/artifacts.md
  • references/guest-insights.md

Open the folder on GitHubat commit 13598cc

Compare with similar skills

Product Taste next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Product Taste compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Product Taste this skillRefoundAI/lenny-skills1.4k—~2kAutomated safety check: PassMIT
Tasteaffaan-m/ECC277k1 repos~3.8kAutomated safety check: PassMIT
Tasteasgeirtj/system_prompts_leaks69k—~506Automated safety check: PassCC0-1.0
Deepseek Reasonruvnet/ruflo74k—~626Automated safety check: NotesMIT
Taste Distillationaffaan-m/ECC277k—~2.3kAutomated safety check: PassMIT
Taste Application Video Pipelineaffaan-m/ECC277k—~4.9kAutomated safety check: PassMIT

Similar skills

  • Taste

    affaan-m/ECC

    Creative-direction layer for music videos and short-form edits in the angelcore / cloud-trance / hyperpop family — a named-genre aesthetic vocabulary, mood + color + light system, beat-synced…

    277k GitHub starsUsed in 1 repo~3.8k tokens
    Media & CreativeAuto-check passed
  • Taste

    asgeirtj/system_prompts_leaks

    Mandatory preflight for web frontend visual design. An agent skill from asgeirtj/system_prompts_leaks.

    69k GitHub stars~506 tokensUpdated yesterday
    Frontend & DesignAuto-check passed
  • Deepseek Reason

    ruvnet/ruflo

    Reasoning-mode completion against DeepSeek's deepseek-reasoner model (R1) via /v1/chat/completions.

    74k GitHub stars~626 tokensUpdated yesterday
    AI & LLM EngineeringAuto-check: notes
  • Taste Distillation

    affaan-m/ECC

    Measure a set of reference videos into a reusable style pack - colour grade as a 3D LUT, cut rhythm as a shot-length distribution, hero stills, screen-blend overlay plates, and a text spec for a…

    277k GitHub stars~2.3k tokensUpdated today
    Auto-check passed
  • Generates video clips against a distilled style pack and cuts them into a finished edit, with grading, overlays, 3D props and numeric checks of the result.

    277k GitHub stars~4.9k tokensUpdated today
    Media & CreativeAuto-check passed
  • Ejentum Reasoning Harness

    sickn33/agentic-awesome-skills

    MCP server exposing four cognitive harness modes (reasoning, code, anti-deception, memory).

    47k GitHub starsUsed in 1 repo~1.9k tokens
    DevelopmentAuto-check passed

More from RefoundAI/lenny-skills

All 76 skills in this repo
  • Customer Interviews

    RefoundAI/lenny-skills

    Help users conduct high-impact customer interviews that move beyond surface-level feature requests to identify root emotional frustrations and specific causal triggers.

    1.4k GitHub stars~1.7k tokensUpdated 2 mo ago
    Auto-check passed
  • Time Energy Management

    RefoundAI/lenny-skills

    Help users master their personal output by shifting from reactive scheduling to intentional energy management, internal trigger mastery, and proactive boundary setting.

    1.4k GitHub stars~1.6k tokensUpdated 2 mo ago
    Auto-check passed
  • User Onboarding Activation

    RefoundAI/lenny-skills

    Help users reach their first moment of core value by optimizing the first-run experience, removing friction, and aligning product design with psychological triggers.

    1.4k GitHub stars~2.1k tokensUpdated 2 mo ago
    Auto-check passed
  • Acquisition Channels

    RefoundAI/lenny-skills

    Help users identify unique distribution advantages and master the lifecycle of acquisition channels to build a sustainable engine for growth and retention.

    1.4k GitHub stars~2.2k tokensUpdated 2 mo ago
    Auto-check passed
  • AI Assisted Prototyping

    RefoundAI/lenny-skills

    Help users build functional product prototypes from natural language or visual mocks using AI coding tools.

    1.4k GitHub stars~1.2k tokensUpdated 2 mo ago
    Auto-check passed
  • AI Evals

    RefoundAI/lenny-skills

    Help users build robust infrastructure for measuring, monitoring, and iterating on AI product performance using human, code-based, and LLM-as-a-judge methodologies.

    1.4k GitHub stars~1.7k tokensUpdated 2 mo ago
    Auto-check passed

Questions about Product Taste

What does Product Taste do?

Help users build a professional eye for quality, articulate the reasoning behind successful product paradigms, and develop the conviction to make high-stakes decisions before data is available. Product Taste is an agent skill from RefoundAI/lenny-skills. Help users build a professional eye for quality, articulate the reasoning behind successful product paradigms, and develop the conviction to make high-stakes decisions before data is available.

How do I install Product Taste in Claude Code?

Run `npx skills add RefoundAI/lenny-skills --skill product-taste -a claude-code`. Or copy the skill folder (skills/product-taste in RefoundAI/lenny-skills) into .claude/skills/product-taste in your project. Claude Code loads it when a task matches its description.

How do I install Product Taste in Codex?

Run `npx skills add RefoundAI/lenny-skills --skill product-taste -a codex`. Or copy the skill folder (skills/product-taste in RefoundAI/lenny-skills) into .agents/skills/product-taste in your project. Codex loads it when a task matches its description.

Can I use Product Taste in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add RefoundAI/lenny-skills --skill product-taste -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/product-taste, .gemini/skills/product-taste, .github/skills/product-taste and .opencode/skills/product-taste in your project.

What does Product Taste need to run?

SKILL.md names no scripts, command-line tools or credentials: Product Taste is instructions for the agent only.

Does Product Taste access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Product Taste safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Product Taste use?

Product Taste is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Product Taste use?

About 2k tokens (SKILL.md is roughly 8.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 24k tokens, read only when the agent opens those files.

What are the alternatives to Product Taste?

Skills that share tags, products or a category with Product Taste: Taste (affaan-m/ECC, 277k stars), Taste (asgeirtj/system_prompts_leaks, 69k stars), Deepseek Reason (ruvnet/ruflo, 74k stars) and Taste Distillation (affaan-m/ECC, 277k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Product Taste?

RefoundAI (a GitHub organization) maintains it in RefoundAI/lenny-skills, which has 1,382 GitHub stars. The repository holds 76 skills in this directory. The repository was last updated on July 16, 2026.

Source: RefoundAI/lenny-skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.