Guide · Agent skill
How to use the Before You Build agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
Before You Build is an agent skill: a set of instructions that tells an AI agent to pause and check an idea before it starts implementing a landing page, MVP, SaaS product, internal tool, agent workflow or major feature. It is written for founders, product managers and AI-assisted builders who want to test demand and positioning before committing engineering time. After this guide you will know what the skill does, how to install it, how to trigger it with a normal request, and where its limits are.
Before You Buildwshobson/agents · MIT Download the skill Quick start
- 1Download and unzip. Download the ZIP from this page and unzip the before-you-build folder.
- 2Place it in a skills folder. Put the folder in ~/.claude/skills/ for personal use, or in the project's .claude/skills/ folder for one project.
- 3Or upload it in the Claude apps. In the Claude apps, go to Settings → Capabilities → Skills and upload the skill.
- 4Describe the build you are about to start. Tell the agent what you plan to build, such as a landing page, MVP or major feature, and who it is for.
- 5Read the five-part output. Check the risk verdict, main assumption, evidence to find first, the do-next step and what to delay.
- 6Run the smallest validation step. Gather the evidence the skill names before writing implementation code, then rerun the review if the picture changes.
What the Before You Build skill makes the agent do
The skill runs a compact pre-mortem before implementation starts. It reviews the idea across seven risks: demand, positioning, monetization, retention, trust, distribution and feature adoption.
The output is short and decision-oriented. The agent names a risk verdict (low, medium or high) with one sentence of reasoning, the main assumption most likely to break the project, the smallest useful evidence to find first, one concrete validation step or reduced build scope, and what to delay building.
The skill is a thinking aid written as instructions. It does not go out and gather market data, so its judgement is only as good as the information you give it.
Install the skill with a ZIP or the Claude apps
Download the ZIP from this page and unzip it. For a personal setup, place the unzipped folder in ~/.claude/skills/. For a single project, place it in that project's .claude/skills/ folder.
If you use the Claude apps, you can instead upload the skill under Settings → Capabilities → Skills.
- Personal install: unzip into ~/.claude/skills/
- Project install: unzip into the project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload
Trigger the skill with natural requests
You do not need a special command. The skill's description tells the agent to use it when you are about to build a landing page, MVP, SaaS product, internal tool, agent workflow or major feature and need to check demand, positioning, monetization, retention, trust, distribution or adoption risk before implementation starts.
Mention that you are about to build something and that you want it checked first. A request such as the one below fits the skill's purpose, because it describes a build and asks for a risk review before coding begins.
- Before I build a landing page for my invoicing tool, check the demand and positioning risks.
- I want to start an MVP for a scheduling app for dog walkers. Review the risks before we write any code.
- We're about to build a major feature for our internal reporting tool. Is the adoption risk worth checking first?
Know when to skip it
The skill is built to stay out of the way for narrow implementation work. Its own instructions say to skip it for a narrow implementation fix, refactor, test repair, dependency update, or an already-validated change with clear acceptance criteria.
Use it for new products, prototypes, landing pages and major features with unclear adoption, revenue, retention, trust or distribution impact. For a bug fix, skip it and let the agent go straight to the work.
Safety notes: scripts and network access
According to the listing, this skill does not run scripts and does not use the network. It is made of instructions in SKILL.md and a reference file, references/risk-checklist.md, which the agent consults when the basic verdict is not obvious.
Because it does not check real market evidence, do not treat its verdict as a final go or no-go decision. Use it to sharpen questions and choose a small validation step.
- Runs scripts: no
- Uses network: no
- Requires: nothing listed
Tips for getting useful output
Give the agent concrete facts about your idea: who the user is, what they do today, and what they currently pay for or use instead. The skill asks for evidence, so vague input produces a vague verdict.
If the verdict is unclear, ask the agent to use the deeper questions in references/risk-checklist.md. Where facts are missing, the skill is designed to name the missing evidence rather than invent market claims, so treat those gaps as your next research task.
- Describe the target user and the workaround they use now
- State whether the idea is already validated and what evidence you have
- Ask for the smallest validation step, not a full research plan
What people use it for
- Checking positioning for a landing page before spending on design or promotion
- Pressure-testing an MVP idea for a founder before committing engineering time
- Reviewing whether a major feature will change user behavior or just add surface area
- Scoping a reduced first version of an agent workflow before building the full system
- Deciding what to delay in a public launch plan until positioning is tested
Frequently asked questions
Is the Before You Build skill free?
The listing states the skill is released under the MIT licence. Check the licence file in the source repository for the exact terms before you redistribute it.
What do I need to use Before You Build?
The listing lists no requirements. It does not run scripts and does not use the network, so it is made of instructions the agent reads.
How do I install the before-you-build skill?
Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use or a project's .claude/skills/ folder. In the Claude apps you can upload it under Settings → Capabilities → Skills.
Why won't the agent use the skill for my bug fix?
The skill is designed to skip narrow implementation work such as bug fixes, refactors, test repairs and dependency updates. Its own instructions say it adds delay without value for those tasks.
Can Before You Build tell me if my product will succeed?
No. It is a thinking aid and does not gather real market evidence. Its judgement is based only on what you tell the agent, so use it to find your riskiest assumption rather than as a final go or no-go decision.
How does this compare with doing market research myself?
The skill does not replace research. It structures the questions and names the smallest validation step, and it says when evidence is missing. You still need to collect the actual signals yourself.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you are a founder, product manager or AI-assisted builder who tends to start coding before checking whether anyone wants the product, or who is planning a public launch. Skip it for bug fixes, refactors and other narrow implementation work, and do not rely on it alone for a final go or no-go decision, since it checks no real market data.
Download the skill