Code craft

Skill mqmalagris/agent-skills/skills/code-craft

Language- and framework-specific code best practices and idioms — complementary to compass (stack-agnostic architecture coach). Two modes: reader (rule + reason + example for a topic) and reviewer (apply rules to user code, flag violations with fixes). Ships rules for TypeScript, Rust, CSS/SASS, Tailwind, Dart, Go, Python and frameworks React, Next.js, RN+Expo, Hono, Cloudflare Workers, Astro, Svelte, Drizzle, Supabase, AWS Lambda+SAM, Terraform, Flutter. Use when the user asks for language-specific best practices ("TS best practices", "is this idiomatic Rust", "/code-craft ts"), wants a code review focused on language idioms (not architecture — see compass), or asks how to do X the right way in a specific language/framework.From its SKILL.md

Install

npx -y skills add mqmalagris/agent-skills --skill code-craft

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

0 stars0 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

6.4 KB, as published. Nobody here has run it

code-craft

Language- and framework-specific code best practices and idiom reference. Complements compass (stack-agnostic architecture / design coach) — code-craft is tactical and per-language.

Quick start

Detect the language or framework from the user's request — explicit (ts, dart, flutter) or inferred from file extension / pasted code.
Detect the mode:
- reader — user asks "what is the best practice for X" or "how should I do Y in Z" → return matching rules from the language file (rule + reason + example).
- reviewer — user pastes code or asks "review this" → load the relevant language file, scan for violations, report location → rule → fix.
Load only the relevant file from languages/ (and a frameworks/ file if applicable).
Match the topic the user asked about — do not dump the entire language file when the user asked one question.

Supported

Language / framework	File	Status
TypeScript	languages/ts.md	shipped
Rust	languages/rust.md	shipped
CSS / SASS	languages/css-sass.md	shipped
React (19)	frameworks/react.md	shipped
Next.js (14 / 15)	frameworks/nextjs.md	shipped
React Native + Expo (SDK 53/54+)	frameworks/react-native-expo.md	shipped
Hono (4.x)	frameworks/hono.md	shipped
Cloudflare Workers	frameworks/cloudflare-workers.md	shipped
Astro (4 / 5 / 6)	frameworks/astro.md	shipped
Svelte 5	frameworks/svelte.md	shipped
Drizzle ORM	frameworks/drizzle.md	shipped
Tailwind CSS (v4)	languages/tailwind.md	shipped
Supabase	frameworks/supabase.md	shipped
AWS Lambda + SAM	frameworks/aws-lambda-sam.md	shipped
Terraform / OpenTofu	frameworks/terraform.md	shipped
Dart 3	languages/dart.md	shipped
Flutter (3.27+)	frameworks/flutter.md	shipped
Go (1.21+)	languages/go.md	shipped
Python (3.12+)	languages/python.md	shipped
Angular, Vue 2, Ionic, Express, Strapi, LoopBack, BigCommerce, Miva	`frameworks/*.md`	legacy / on demand

If the user asks for a language not yet supported, say so and offer to add a stub via the structure documented below.

When to defer to compass

Defer to compass if the user asks about:

Architecture / system design
SOLID, GoF patterns, MVC, microservices
Refactoring strategy across modules
TDD / CI/CD / DevOps
Stack selection or technology trade-offs
Security model, threat surface, perf budget

code-craft is for inside-the-file decisions. compass is for across-file / across-system decisions. When a user request straddles both, route the language-idiom parts here and the architecture parts to compass.

Reader-mode output format

For a single-rule ask:

Rule. Reason. Wrong vs. right example (3–6 lines).

For a topic ask (e.g. "TS error handling"), return the 3–6 most relevant rules in the same compact format, ordered by importance. Cite the source when there is genuine community debate (enums, barrel files, etc.).

Reviewer-mode output format

Flat numbered list, one violation per line:

1. <file>:<line> — <rule short name> — <one-sentence fix>
2. ...

If the list is long, group by severity:

must-fix — correctness or safety bugs
should-fix — idiom or maintainability
nice-to-have — style only

Close with a one-sentence summary. Do not re-print the user's code.

Adding a new language or framework file

Create languages/<lang>.md or frameworks/<framework>.md.
Group rules under buckets that fit the language — typical: tactical (day-to-day), idioms (ecosystem patterns), anti-patterns (smells).
Each rule:
- Short name (1–6 words)
- One-sentence rule
- One-line reason
- ≤ 6-line wrong + right example
- Source citation if there is debate
Update the Supported table above.
Aim for 30–60 rules per language file. Quality over quantity.

Confirmation discipline

Reader mode: no confirmation needed; just answer.
Reviewer mode: confirm scope before scanning ("idioms only, or include anti-patterns?"). Default to both. Security and performance reviews belong in compass — route there if asked.

When to skip code-craft

One-line tweaks (rename, fix typo, add import).
Throwaway scripts / spikes.
Generated code (migrations, codegen output, build artifacts).
Language not in the Supported table — say so, don't fabricate rules from training data.
User asks about architecture, security, or perf — route to compass instead.

Pipeline placement

grill-me → to-prd → compass → heist → maestro → code (+ code-craft) → compass (review)

code-craft sits during the code phase, applied tactically per-file as code lands. Use cases:

Drafting — when writing new code in a supported stack, apply rules silently to the just-written diff and self-correct before showing it.
PR review — alongside compass's reviewer workflow, run code-craft on the diff for stack-specific idiom violations.

code-craft does NOT replace compass's reviewer workflow. They run side-by-side at PR time: compass for cross-file architecture, code-craft for inside-file idioms.