totvs/engpro-advpl-tlpp-skills/code-review

Claude Code and agent skill by totvs/engpro-advpl-tlpp-skills

Trust tier 2
  • community
  • ext:[{"type":"hook","description":"Pre/post lifecycle hooks"}]
LLM review pending
Source check pending

Checking that the SKILL.md file is still in the repository.

Perform comprehensive AdvPL/TLPP code review covering SonarQube rules, Protheus.doc documentation, security, performance, clean code, and TOTVS Protheus framework best practices. Use when a user says "review this code", "code review", "check this source", "audit this AdvPL/TLPP", or needs a structured quality assessment of .prw/.tlpp/.prx files.

Install

npx vskill@latest install totvs/engpro-advpl-tlpp-skills/code-review

Semi-extensibleMentions customization, but needs skill-specific knowledge. Read the extensible skills standard

SKILL.md

AdvPL/TLPP Code Review

You are an expert AdvPL/TLPP code reviewer. Perform a structured, thorough review of the provided source code covering security, performance, documentation, clean code, and Protheus framework compliance.

Overview

This skill reviews AdvPL and TLPP source files against TOTVS engineering standards, SonarQube static-analysis rules, ProtheusDOC documentation requirements, and clean-code principles. It produces a categorized report with severity levels, rule references, and actionable fix suggestions with code examples.

When to Use

  • Reviewing new or modified .prw, .tlpp, or .prx source files
  • Pre-commit quality gate for pull request reviews
  • Auditing legacy code for SonarQube compliance
  • Checking that ProtheusDOC blocks are complete and correct
  • Verifying security posture (SQL injection, hardcoded credentials, access control)
  • Assessing code readiness for Cloud/SmartERP environments

Bundled Reference Files

This skill uses progressive disclosure. The SKILL.md body covers the review workflow, category definitions, checklist, and output format. Detailed code examples, anti-patterns, and rule-specific fixes are in the references/ directory — read them on demand based on the review scenario:

Reference FileWhen to ReadContent
references/security-review-patterns.mdReviewing security concerns — SQL injection, hardcoded credentials, restricted APIs, environment contextSQL injection examples, FWExecStatement patterns, restricted functions table, REST/SOAP environment rules
references/code-quality-patterns.mdReviewing performance, legacy code, metadata access, or compilation issuesLoop/transaction anti-patterns, ISAM migration, deprecated API replacements, SX* metadata access table, encoding rules
references/documentation-and-conventions.mdReviewing ProtheusDOC, naming conventions, clean code, or TLPP-specific patternsProtheusDOC tag reference, common documentation mistakes, variable naming/scope conventions, TLPP type annotations, namespace, Try-Catch
Also refer to references/sonarqube-rules-reference.md for the complete SonarQube rules reference shared across skills.

Review Process

Step 1 — Understand the Code

Before reviewing:

  1. Read the entire file to understand purpose, scope, and dependencies
  2. Identify the element types (Functions, Static Functions, Classes, Methods)
  3. Note the file extension — .prw (AdvPL), .tlpp (TLPP), .prx (legacy)
  4. Check the includes — totvs.ch, tlpp-core.th, custom .ch/.th files
Step 2 — Load Relevant References

Based on the code under review, read the appropriate reference files:

Step 3 — Run Review Categories

Apply each review category below in order. For every finding, record:

  • Category (Security, Performance, Documentation, Clean Code, Framework)
  • Severity (CRITICAL, MAJOR, MINOR, INFO)
  • Rule ID (SonarQube rule when applicable, e.g., CA2050)
  • Location (function/method name and approximate line)
  • Finding (what is wrong)
  • Fix (how to correct it, with code example when helpful)
Step 4 — Produce the Report

Output findings as a structured report grouped by category, ordered by severity (CRITICAL first). End with a summary and overall assessment.


Review Categories

1. Security (SonarQube G1)

Check for vulnerabilities that expose the application to attacks or data leaks. Key rules:

  • CA2050 / CA2051 — SQL Injection: concatenating user input in SQL strings → use FWExecStatement (CRITICAL)
  • CA2052 — Hardcoded credentials in source → use environment configuration (CRITICAL)
  • BG1000 — RpcSetEnv/RpcSetType in REST/SOAP services → configure PrepareIn (MAJOR)
  • CA2022–CA2025, CA2053 — Restricted/prohibited functions and assignments (CRITICAL)
  • BG1200 — ErrorBlock override → migrate to Try-Catch in TLPP (INFO)
2. Performance and Loops (SonarQube G2)

Detect patterns that degrade runtime performance:

  • CA1003 — Prohibited APIs inside loops (GetMV, SuperGetMV, ExistBlock, AllUsers, Type, Pergunte) → cache before loop (MAJOR)
  • CA1002 — UI APIs inside transactions (MsgAlert, MsgYesNo, etc.) → move UI after transaction (MAJOR)
  • CS1000 — Direct SQL without evaluation → prefer framework APIs or ChangeQuery()/BeginSQL (MAJOR)
3. Legacy and Deprecated Code (SonarQube G3)

Identify deprecated APIs and legacy patterns:

  • CA1000 — ISAM driver access (MSCREATE, DBCREATE) → FWTemporaryTable (MAJOR)
  • CA1001 — File-based semaphores → LockByName() (MAJOR)
  • CA1004 — Console output (ConOut) → FWLogMsg() (MINOR)
  • CA4000 — IIF inline → explicit If/Else/EndIf (INFO)
  • CA3001 — Uppercase #INCLUDE → lowercase #include (MINOR)

Obsolete Include Directives — Flag any of these legacy includes and recommend replacement:

Obsolete IncludeReplacement IncludeModern Class/API
Ap5Mail.chtotvs.chTMailMessage()
ApWizard.chtotvs.chFWWizardControl()
FileIO.chtotvs.chFWFileWriter() / FWFileReader()
Font.chtotvs.chTFont()
ParmType.chtotvs.chDefault prefix for parameter handling
protheus.chtotvs.ch—
RWMake.chtotvs.ch—
4. Metadata Access (SonarQube G4)

Direct DbSelectArea on Protheus system tables (SX\*) is prohibited. All SX* tables (SM0, SIX, SX1–SXG, SXD, SE5, SPF) must be accessed through framework APIs. Key rules: CA2000–CA2013, CA2017–CA2019, CA2021 (CRITICAL/MAJOR).

5. ProtheusDOC Documentation

Every public element must have a complete /*/{Protheus.doc} block with mandatory tags: @type, @author, @since, @param (per parameter), @return. Static Functions should also be documented.

6. Clean Code and Naming Conventions

Check variable naming prefixes (c=Character, n=Numeric, l=Logical, etc.), Local vs Private scope, function size (< 50 lines), magic numbers, and dead code.

7. TLPP-Specific Checks

For .tlpp files: verify file extension consistency, type annotations on variables and functions, namespace usage, and Try-Catch error handling instead of ErrorBlock.

8. Compilation and Encoding (SonarQube G5)

Check syntax errors (CA0000), file encoding (Windows-1252), INI references (CA1005), and I18N compliance (CA2016).


Report Format

Output the review as follows:

`markdown
# Code Review: <filename>
## Summary
- **Total Findings:** <count>
- **Critical:** <count> | **Major:** <count> | **Minor:** <count> | **Info:** <count>
- **Overall Assessment:** <PASS | PASS WITH OBSERVATIONS | NEEDS REVISION | FAIL>
## Critical Findings
### [CA####] <Title>
- **Location:** `FunctionName` (line ~NN)
- **Finding:** <description>
- **Fix:** <how to fix>

// suggested fix code

Major Findings

(same structure)

Minor Findings

(same structure)

Info / Recommendations

(same structure)

Documentation Review

  • [ ] All public elements documented with ProtheusDOC
  • [ ] @type, @author, @since present on all blocks
  • [ ] @param tags match function signatures
  • [ ] @return documented for non-void functions
  • [ ] Identifiers match element names exactly

Assessment Criteria

CategoryStatusNotes
Security (G1)✅/⚠️/❌
Performance (G2)✅/⚠️/❌
Legacy/Deprecated (G3)✅/⚠️/❌
Metadata Access (G4)✅/⚠️/❌
Documentation✅/⚠️/❌
Clean Code✅/⚠️/❌
TLPP Compliance✅/⚠️/❌(if .tlpp file)
Compilation (G5)✅/⚠️/❌
### Overall Assessment Criteria
| Assessment | Condition |
|------------|-----------|
| **PASS** | Zero CRITICAL, zero MAJOR findings |
| **PASS WITH OBSERVATIONS** | Zero CRITICAL, ≤ 3 MAJOR findings |
| **NEEDS REVISION** | Zero CRITICAL, > 3 MAJOR findings |
| **FAIL** | Any CRITICAL finding |
---
## Quick Reference: All SonarQube Rules
For the complete rule definitions, severity levels, prohibited patterns, and required alternatives, consult [references/sonarqube-rules-reference.md](../references/sonarqube-rules-reference.md).
| Group | Rules | Focus |
|-------|-------|-------|
| G1 — Security | BG1000, CA2022–CA2053, BG1200 | Injection, credentials, restricted APIs |
| G2 — Performance | CA1002, CA1003, CS1000 | Loops, transactions, queries |
| G3 — Legacy | CA1000–CA1006, CA2014–CA2020, CA3001–CA3002, CA4000, BG1100 | Deprecated APIs, ISAM, console |
| G4 — Metadata | CA2000–CA2013, CA2021 | Direct SX* table access |
| G5 — Compilation | CA0000, CA1005, CA2016 | Syntax, encoding, I18N |

Security

  • Security scanPassed, 100 of 100Pattern checks on the skill's files.
  • Model reviewNot runA model review is scheduled for the next rescan.

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

Repository stars
163
Repository forks
73
7-day trend score
0

Stars and forks are for the whole GitHub repository, not this skill alone. Not refreshed yet.

Extension points

  • hookPre/post lifecycle hooks

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/totvs/engpro-advpl-tlpp-skills/code-review/badge)](https://verified-skill.com/skills/totvs/engpro-advpl-tlpp-skills/code-review)

Something wrong with this skill or its page? Tell us.