From coding-standards
Single source of truth for all coding rules, conventions, and quality standards. Auto-loaded at conversation start. Sub-files loaded on-demand based on the work being done.
How this skill is triggered — by the user, by Claude, or both
Slash command
/coding-standards:coding-standardsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
This is the authoritative reference for how code should be written across all projects. When rules conflict between files, this skill wins.
checklists/before-committing.mdchecklists/before-creating.mdlint-config.mdrules/component-architecture.mdrules/convex-backend.mdrules/error-handling.mdrules/file-organization.mdrules/general-quality.mdrules/naming-conventions.mdrules/nodejs-backend.mdrules/react-patterns.mdrules/reuse-first.mdrules/security.mdrules/state-management.mdrules/tailwind-and-tokens.mdrules/types-and-constants.mdrules/typescript-quality.mdThis is the authoritative reference for how code should be written across all projects. When rules conflict between files, this skill wins.
SKILL.md (this file) is auto-loaded every conversation. It gives Claude the philosophy and a manifest of sub-files. The sub-files are read on-demand based on what you're working on.
| File | Read when... |
|---|---|
rules/reuse-first.md | Creating ANY new component, type, constant, or hook |
rules/component-architecture.md | Building or modifying React components |
rules/naming-conventions.md | Naming anything (files, components, variables, functions) |
rules/file-organization.md | Creating files, moving files, structuring directories |
rules/types-and-constants.md | Creating or extracting types, interfaces, or constants |
rules/typescript-quality.md | Writing TypeScript (always) |
rules/react-patterns.md | Working with React/Next.js components, hooks, pages |
rules/tailwind-and-tokens.md | Writing className, styling, using design tokens |
rules/convex-backend.md | Working in the convex/ directory |
rules/nodejs-backend.md | Server Actions, API routes, env vars, logging, dates |
rules/security.md | Auth, validation, user input, secrets, external data |
rules/error-handling.md | Error boundaries, retries, resilience, error propagation |
rules/state-management.md | Working with state (Zustand, URL params, server state, forms) |
rules/general-quality.md | Always — JS idioms, comments, magic values, defensive coding |
| File | Use when... |
|---|---|
checklists/before-creating.md | BEFORE creating any new file (component, hook, type, constant) |
checklists/before-committing.md | AFTER writing code, before committing |
lint-config.md defines which rules are BLOCKING (must fix) vs WARNING (should fix). The /lint skill references this for severity levels.
These rules apply to EVERY file, EVERY time:
any — use unknown and narrowexport default — named exports only (except page/layout)index.ts that re-exportscn() for all conditional classes — never template literalsctx.db.delete(), always deletedAtchecklists/before-creating.md firstGuides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Enforces test-driven development: write failing test first, then minimal code to pass. Use when implementing features or bugfixes.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.
npx claudepluginhub likeahuman-ai/likeahuman --plugin coding-standards