agentsclimarketplace
Sign in

Scaffold

Skill Brayanhenaor/ai-standards/skills/scaffold

Generate a complete feature scaffold across the project's layers, following its real conventions. Use when starting a new feature, endpoint, or module and you want the boilerplate laid down consistently. Mirrors the existing architecture instead of imposing one, and scaffolds only the layers the project actually uses.From its SKILL.md

Install
npx -y skills add Brayanhenaor/ai-standards --skill scaffold

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.6 KB, as published. Nobody here has run it

Scaffold

Lay down a new feature the way this project already does it — same layers, same patterns, same conventions — so the new code looks like it was always there.

1. Learn the project's shape first

  • Read an existing comparable feature end to end. That's the template, not a generic blueprint.
  • Identify the real architecture (layered, vertical slices, single project…), naming, folder layout, error-handling and validation approach, and how requests flow from entry point to data.
  • Use the project profile from init/CLAUDE.md if present. Don't assume Clean Architecture or any pattern the project doesn't use.
  • Confirm before generating. Present the proposed file list and layer shape and check with the developer first — scaffolding writes many files and picks an architecture; don't lay down a skeleton they haven't agreed to.

2. Generate, mirroring conventions

  • Produce the files the feature needs across the layers the project actually has — entry point (endpoint/controller/handler), application logic, domain types, data access, DTOs, registration — named and placed like the existing code.
  • Apply clean-code-core and the active pack: inject abstractions, thin entry points, validation at the boundary, DTOs not entities, the project's result/error pattern.
  • Don't hand-roll the hard parts. Before scaffolding a non-trivial piece (auth, validation, mapping, parsing, file handling), apply anti-overengineering: if the platform or a validated, commercially-free library already solves it, propose that (license + consent) instead of generating it from scratch.
  • Scaffold only what's needed. Don't generate layers, interfaces, or abstractions the feature doesn't use just to fill a template — that's over-engineering. A simple feature gets a simple scaffold.

3. Include tests and wiring

  • Add tests matching the project's test conventions (see the test skill) for the real paths.
  • Wire up DI registration, routing, and configuration so it compiles and runs — not orphan files.

4. Verify and hand off

Confirm it builds (verify). Summarize what was created and what the developer still needs to fill in (business logic, real validations). Flag any decision worth a plan or ADR. Reply in the developer's language; code and identifiers in English.

Keep looking

Skills are one crate of 627,801. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.