Code Review Standards

Review checklist

Correctness and logic

• Logic is correct and edge cases are handled.
• No obvious bugs or race conditions.
• Error handling is present and consistent.

Security

• No keys, tokens, or secrets in code; use

  .env

  or environment variables.
• User input is sanitized; no raw SQL (use placeholders).
• Pydantic or equivalent used for input validation.

Code standards

• Type hints and docstrings (Google style) on public functions.
• snake_case for variables/functions/files; PascalCase for classes; UPPER_CASE for constants.
• Functions under ~40 lines; DRY and SRP respected.
• f-strings and logging (no print for production).
• Imports at top; no inline imports.

Tests and quality

• Unit tests for new/changed behavior;

  tests/

  with

  test_<module>.py

  .
• Target 80% coverage for touched code.
• Version checks:

  VERSION

  exists; UI or

  /version

  shows "versão:" (or equivalent).

Architecture and delivery

• Clean separation (domain, use cases, infrastructure); no business logic in controllers.
• Repository / Service Layer where applicable; no circular dependencies.

• funcoes.md

  updated if functions were added, changed, or removed.
• Code formatted (e.g. Black, Ruff); Conventional Commits if applicable.

DevOps and config

• Dockerfile and docker-compose (and app-only variant) if the project uses Docker.

• .env.example

  documents required variables.
• No global deps or cache committed.

Feedback format

• Critical: Must fix before merge (security, correctness, blocking).
• Suggestion: Should improve (readability, performance, consistency).
• Nice to have: Optional improvement.