Claude-skill-registry claude-framework

CLAUDE Framework coding standards for code quality, naming conventions, error handling, security, testing, database, and logging

install
source · Clone the upstream repo
git clone https://github.com/majiayu000/claude-skill-registry
Claude Code · Install into ~/.claude/skills/
T=$(mktemp -d) && git clone --depth=1 https://github.com/majiayu000/claude-skill-registry "$T" && mkdir -p ~/.claude/skills && cp -r "$T/skills/data/claude-framework" ~/.claude/skills/majiayu000-claude-skill-registry-claude-framework && rm -rf "$T"
manifest: skills/data/claude-framework/SKILL.md
source content

CLAUDE Framework

Production-ready coding standards for building maintainable, secure, and well-tested software.


Code Quality (C-1 to C-5)

  • C-1: Single Responsibility Principle - each function/class does ONE thing
  • C-2: DRY (Don't Repeat Yourself) - no code duplication
  • C-3: KISS (Keep It Simple) - simplicity over complexity
  • C-4: Functions maximum 20 lines (split if longer)
  • C-5: Prefer composition over inheritance

Naming Conventions (N-1 to N-6)

  • N-1: Use descriptive names that explain intent
  • N-2: Functions = verbs:
    calculateTotal()
    ,
    validateUserInput()
  • N-3: Variables = nouns:
    userAccount
    ,
    totalPrice
  • N-4: Booleans start with is/has/can/should:
    isValid
    ,
    hasPermission
  • N-5: Constants in UPPER_SNAKE_CASE:
    MAX_RETRY_ATTEMPTS
  • N-6: Avoid abbreviations: use
    user
    not
    usr

Error Handling (E-1 to E-5)

  • E-1: Handle ALL possible error scenarios
  • E-2: Use specific error types/messages
  • E-3: Log errors with context information
  • E-4: NEVER allow silent failures
  • E-5: Fail fast - validate inputs early

Security (SEC-1 to SEC-5)

  • SEC-1: Validate ALL inputs at system boundaries
  • SEC-2: Sanitize output data
  • SEC-3: Use environment variables for secrets
  • SEC-4: Never hardcode sensitive information
  • SEC-5: Implement proper authentication and authorization

Testing Standards (T-1 to T-5)

  • T-1: Write failing test first, then implement (TDD)
  • T-2: Minimum 80% code coverage for new code
  • T-3: Test happy path, error scenarios, and edge cases
  • T-4: Descriptive test names explaining what is tested
  • T-5: Arrange-Act-Assert pattern clearly separated

Test Quality (TQ-1 to TQ-5)

  • TQ-1: Use realistic test data, no magic numbers
  • TQ-2: One assertion per test where possible
  • TQ-3: Ensure test isolation
  • TQ-4: Tests must be deterministic (no flaky tests)
  • TQ-5: Tests should run fast (mock external dependencies)

Database (DB-1 to DB-4)

  • DB-1: Use transactions for multi-step operations
  • DB-2: Optimize queries (avoid N+1 problems)
  • DB-3: Document indexing strategy
  • DB-4: Create migration and rollback scripts

Logging (L-1 to L-4)

  • L-1: Structured logging (JSON format)
  • L-2: NEVER log sensitive data
  • L-3: Include correlation IDs for tracing
  • L-4: Use appropriate log levels: DEBUG, INFO, WARN, ERROR

Code Structure Requirements

  • Organize imports clearly (stdlib, third-party, local)
  • Define constants at module level
  • Use pure functions where possible
  • Implement proper error boundaries
  • Include comprehensive JSDoc/comments for complex logic
  • Follow consistent indentation (2 or 4 spaces)
  • Maximum 120 characters per line

Quality Assurance Checklist

Before delivering code, verify:

  • ✅ Functions under 20 lines
  • ✅ Single responsibility maintained
  • ✅ No code duplication
  • ✅ Clear naming conventions followed
  • ✅ All errors handled with specific types
  • ✅ Input validation implemented
  • ✅ Output sanitization applied
  • ✅ Tests written and passing (80%+ coverage)
  • ✅ Security considerations addressed
  • ✅ Performance implications considered
  • ✅ Documentation updated