Menu
Review product and feature risk before an AI coding agent starts implementation.
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/ComeOnOliver/skillshub --skill before-you-buildThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Showing SKILL.md
Use Xquik for X data and confirmation-gated X actions: tweet search, user lookup, follower export, media download, monitors, webhooks, MCP, and SDK workflows.
Canton Network open-source ecosystem guide covering DAML SDK, Canton runtime, and Splice applications. Use when working with Canton Network, DAML smart contracts, or building decentralized applications.
Manages Linear issues, teams, and projects via CLI. Lists issues, creates tasks, views details, links issues, and runs GraphQL queries. Must use for "my Linear issues", "create Linear task", "link issues in Linear", "Linear API query", or any Linear project management request.
Based on SOC occupation classification
| name | before-you-build |
| description | Review product and feature risk before an AI coding agent starts implementation. |
Don't ask AI to build it yet. Ask why it might fail first.
Use this skill before building or changing a product, feature, SaaS, AI app, side project, or startup idea.
Use it when the user asks:
No API keys or external services are required.
If live evidence would improve the review, ask the user whether they want you to search for similar products, competitor complaints, communities, or demand signals.
First, do not write code, scaffold a project, recommend a tech stack, or create an implementation plan.
Review whether the idea should move into implementation at all.
If the idea is too broad, ask the user to complete this sentence:
This tool is for [specific people], in [specific situation], to solve [specific problem].
If the current alternative is missing and the review would be too speculative, ask:
How do they solve this today, and why is that not good enough?
Ask at most two questions before giving a constrained review.
Then produce a short Quick Reality Check:
## Quick Reality Check
Assumption:
- [State the assumption if any.]
Verdict:
- Don't build yet / Build smaller / Build only if / Build small
Biggest risk:
- [The most important likely failure mode.]
Most likely problem:
- [Demand / distribution / pricing / positioning / retention / trust.]
What to validate first:
- [One concrete test before implementation.]
Smallest useful version:
- [A narrower version worth testing.]
For a new product idea, check whether the user has a narrow buyer, specific situation, painful problem, current alternative, and reachable distribution channel.
For a feature request, check whether it comes from repeated demand, a paying segment, a retention blocker, or only a loud edge case.
For a requirement change or pivot, check whether the change improves validated demand or only expands scope.
For a learning or portfolio project, do not judge by startup standards. Keep the scope small and focus on the learning artifact.
Do not reward vague AI app ideas with implementation plans.
Do not treat "a competitor exists" as proof of demand.
Do not treat friends saying "cool idea" as validation.
Do not turn the review into a long questionnaire.
Do not block all building. The useful outcome may be "build smaller" or "validate this narrow version first."
Full project: https://github.com/bin1874/before-you-build-skill