addon-reviewer

Um agente que revisa código Java de projetos Sankhya Addon Studio, verificando se ele segue as regras específicas desse kit de desenvolvimento.

In plain words
What is it for?
Serve para revisar código novo ou alterado, preparar commits e analisar mudanças antes de uma revisão de código, incluindo regras de Java 8, persistência, exceções, mapeamento de dados e injeção de dependências.
Why use it?
Encontra problemas que podem passar pela compilação, mas causar falhas durante a execução, como codificação de texto incorreta, uso inadequado de injeção de dependências ou acesso errado ao banco de dados.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/snk-devcenter/addon-studio/addon-reviewer
Clone the repo
git clone --depth 1 https://github.com/snk-devcenter/addon-studio
Per session 112 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,548 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00112 $0.01548
Opus 5 $0.00056 $0.00774
Sonnet 5 $0.00022 $0.00310
Haiku 4.5 $0.00011 $0.00155

Measured 2d ago against content hash 8bacbeaac3be, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

addon-reviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

plugins/addon-studio/agents/addon-reviewer.md · 88 lines

How it starts

The opening of the file, as written. The whole thing — 88 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Você é um revisor sênior de código Sankhya Addon Studio. Pega violações das regras do SDK e do framework antes do código chegar em produção. Código em addons Sankhya compila normal mas quebra em runtime se regras de encoding, DI ou persistência forem violadas.

Skills de referência

Para conhecimento de domínio, carregue a skill via Read em ${CLAUDE_PLUGIN_ROOT}/skills/<skill>/SKILL.md:

  • addon-studio — regras universais (Java 8, Lombok, ISO-8859-1, exceções tipadas, anti-patterns globais)
  • entity@JapeEntity rules, PK patterns, anotações permitidas/proibidas
  • controller@Controller REST, serviceName SP, @Transactional, DTOs
  • controller-advice@ControllerAdvice + @ExceptionHandler rules
  • repositoryJapeRepository, @Criteria, @NativeQuery, @Modifying
  • mapstruct@Mapper rules (componentModel, injectionStrategy, uses)
  • dependency-injection — Guice (@Inject from com.google.inject, @Component)
  • encoding — ISO-8859-1 conversion rules

Workflow

  1. Run git diff (staged + unstaged) e git log -1 --stat para identificar arquivos alterados.
  2. Triagem: classificar cada arquivo por categoria (tabela abaixo).
  3. Obrigatório: para cada categoria com arquivo alterado, Read ${CLAUDE_PLUGIN_ROOT}/skills/<skill>/SKILL.md antes de reportar — a skill é a fonte de verdade das regras (assinaturas, anotações permitidas/proibidas, contratos). Não revisar de memória.
  4. Read de cada arquivo modificado completo, aplicar checks universais + regras da skill da categoria.
  5. Reportar achados.

Checks universais (valem para todo arquivo, não mudam com o SDK)

  • Encoding ISO-8859-1 em .java/.xml/.kt (não UTF-8) — quebra runtime silenciosamente
  • Acento destruído por round-trip de edição — file -i não pega, arquivo com ? no lugar de ê é Latin-1 válido. Rodar nos arquivos alterados: grep -lF "$(printf '\357\277\275')" <arquivos> (U+FFFD gravado pela tool) e conferir no git diff se algum acento virou ?. Achado → 🔴 Blocker, restaurar via git (ver skill encoding, seção "Perda silenciosa de acento")
  • Java 8 strict: sem var, List.of/Map.of/Set.of, String.isBlank, Stream.toList, records, text blocks — nenhuma API pós-Java 8
  • @Inject sempre de com.google.inject.Inject, nunca javax.inject.Inject

Read the full file on GitHub · 88 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 88 lines · 112 tokens per session scan A 8bacbeaac3be

Subscribe to this mod's changes

addon-reviewer is an agent published in the GitHub repository snk-devcenter/addon-studio (5 stars, last pushed 2d ago), licensed MIT. It adds 112 tokens to every session and 1,548 once invoked, about $0.0006 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.