THE WANTLIST BUILDER LIBRARY · VOL. 01
A field guide to turning an idea into software you understand, can improve and are ready to share.
48 building blocks and seven recipes, from Next.js to Clerk.
Twelve habits with observable checks and official references.
18 original templates for every stage of the build.
LEARN BY BUILDING
Read in order, or start with the question in front of you.
What vibe coding does well, and where deliberate engineering begins.
Compare app builders and repository agents using the same small task.
Turn an idea into boundaries, examples and acceptance criteria.
Plan, implement, inspect and test one complete behavior at a time.
Next.js, PostgreSQL, authentication, files and an AI gateway.
Reproduce a failure and find its cause before adding another patch.
Separate domain control, functioning features and security review.
Budget for usage, hosting and maintenance; rehearse moving the app.
Try the hosted signup, upload and confirmation journey.
Share a useful change, ask a specific question and measure interest.
CHOOSE YOUR ENVIRONMENT
Start with the kind of work you want to do. Test two tools on the same small task.
10 tools
Prompt → web app
Build a web application from a natural-language description, with editable code and Git sync.
Describe a booking flow, then verify that two users cannot edit each other’s bookings.
Build in the browser
Create and iterate on an application in a browser workspace.
Create a small tracker, reload a saved record and try its deployed URL.
Idea → hosted project
An agent inside Replit that can set up a project, implement changes and work toward deployment.
Build one authenticated workflow and repeat it outside the preview session.
Interface → application
Vercel’s AI building environment for creating interfaces and applications with a live preview.
Build a dashboard from sample data, then replace one panel with a real persisted record.
Work inside your repo
An agent that searches a codebase, edits files and runs commands in a coding environment.
Ask it to trace a failing save operation, fix the cause and show the regression check.
Inspect · edit · run
A coding agent with a CLI for exploring repositories, changing code and running development tools.
Ask for a repository map, then implement one feature with explicit acceptance criteria.
Codebase → tested change
An agent that reads code, edits files and runs commands across terminal, IDE, desktop and web surfaces.
Ask it to inspect an upload path and reproduce one failure before editing.
Issue → reviewed change
Copilot’s cloud agent can work on a repository in an ephemeral environment and prepare a pull request.
Assign a small issue with a failing example and review the resulting pull request.
Editor + terminal
An agent that can read and write files, run commands and use a browser from coding workflows.
Use it to make one interface change, then inspect the result in the browser.
Context-aware editing
Cascade provides code and chat modes, tool use and checkpoints in Devin Desktop.
Trace a feature across files and review a small implementation change.
Capabilities are summarized from official documentation. Fit and trial suggestions are Wantlist’s editorial judgment, not benchmark rankings. Categories overlap. Check current plans, limits and runtime support before committing.
GIVE THE AGENT SOMETHING CONCRETE
Original templates to adapt to your project. Replace the brackets and keep secrets out.
I am building [product] for [specific user]. Their current workaround is [workaround]. The first useful outcome is [outcome]. Draft a brief with: one user journey, in-scope and out-of-scope behavior, data ownership, failure states and five observable acceptance criteria. List assumptions and unresolved decisions. Do not implement yet.
Inspect the existing project and its instructions. Implement [one behavior] using the established stack and patterns. Acceptance criteria: [inputs and expected results]. Include [failure case]. Preserve unrelated work. Run the relevant checks and report what you tested, what remains unverified and how to try the change.
Reproduce this failure: [steps]. Expected: [result]. Actual: [result]. Relevant error, with secrets removed: [error]. Trace the request through the interface, server and storage. State the evidence for the cause before editing. Make a focused fix, then repeat the failing case and the normal path. Report any remaining uncertainty.
In a test environment, create accounts A and B and one private record owned by A. Check list, direct read, update, delete and file access as B and signed out. Use the actual API as well as the UI. Expect every unauthorized access to fail. Record the result and affected route for each test. Remove only your test fixtures. Do not use real customer data.
Document how to run this application from a fresh checkout. Include exact runtime versions, install/build/start commands, required environment variable names without values, migration order, storage dependencies and authentication callbacks. Identify platform-specific assumptions. Test the documented process in an isolated environment and report every step you could not verify.
Review the deployed test version at [URL]. Test this journey: [visitor action → signup → first useful outcome]. Check mobile layout, keyboard use, validation, persistence, account isolation and any confirmation email. For each claim, report evidence, version, date and limitations. Separate passed, failed and untested checks. Do not charge a payment method or contact real users.
PLAIN ENGLISH, PLEASE
Short working definitions, with the practical consequence attached.
20 terms
FOLLOW THE EVIDENCE
Provider documentation for capabilities; original Wantlist explanations for practice. This edition is a reviewed snapshot, not a live ranking.
We avoid copying third-party guides and do not infer app quality from provider claims. Sources support the associated topic, not a blanket endorsement. Pricing and model leaderboards are intentionally omitted because a useful comparison needs a current workload and billing assumptions.
