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.
git clone --depth 1 https://github.com/bittencourtthulio/expxdevWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/bittencourtthulio/expxdev/prodx-avaliar)<a href="https://agentmods.dev/commands/bittencourtthulio/expxdev/prodx-avaliar"><img src="https://agentmods.dev/badge/commands/bittencourtthulio/expxdev/prodx-avaliar/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/bittencourtthulio/expxdev/prodx-avaliar"><img src="https://agentmods.dev/badge/commands/bittencourtthulio/expxdev/prodx-avaliar.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00032 | $0.00621 |
| Opus 5.5 | $0.00013 | $0.00248 |
| Sonnet 5.5 | $0.00006 | $0.00124 |
| Haiku 4.5 | $0.00003 | $0.00062 |
Grade A, and why
prodx-avaliar 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 today.
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.
How it starts
The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Use a skill prodx, etapas P2 a P5.
O pedido está em $ARGUMENTS — PD-ID, ou o texto do pedido. Se vier vazio, liste os pedidos abertos com o estágio de cada um e pergunte qual.
Antes de começar
Confirme que a triagem rodou e que pelo menos um gatilho disparou (regra 3). Se não rodou, rode-a primeiro. Se nenhum gatilho disparou, não abra avaliação completa: dê o veredito direto.
Consulte a máquina de estados do SKILL.md e execute a etapa pendente, não a primeira. Se o usuário pedir uma etapa adiantada, explique o que falta e execute a pendente.
Etapas
| Etapa | Reference | Saída |
|---|---|---|
| P2 — entendimento | references/02-pedido.md |
01-pedido.md |
| P3 — existência | references/03-existencia.md |
02-existencia.md |
| P4 — avaliação | references/04-avaliacao.md |
03-avaliacao.md |
| P5 — veredito | references/05-veredito.md |
VEREDITO.md |
Atalho obrigatório: se o P3 der EXISTE, pule o P4 e vá direto ao P5 com veredito ja_existe.
Pare no P2 se houver pergunta essencial em aberto (regra 8): informação que falta vira pergunta, nunca suposição.
Tom
Modo didático, padrão nesta casa: explique o raciocínio de cada etapa enquanto trabalha, mostre o que considerou e descartou, e diga por que aquela pergunta importa.
Passe ao modo conciso quando as duas condições valerem: signatário com provisorio: false no PRODUTO.md e mais de trinta pedidos avaliados no INDICE.md.
Regras
- Regra 4: a solução pedida nunca é o requisito. O requisito é o problema.
- Regra 5: existência com evidência — tela, rota, arquivo ou relatório.
- Regra 6: escopo mínimo nunca fica vazio.
- Regra 7: pedido rotulado como bug é verificado contra comportamento intencional.
- Regra 1: a skill não assina. Deixe
aprovado_poreaprovado_emcomoPENDENTEe diga quem deve assinar.
Ao terminar
Apresente a recomendação com força, diga por que ela e não as outras três, e peça a assinatura de quem está registrado no PRODUTO.md. Atualize a linha do pedido no INDICE.md.
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.
- today First seen · 45 lines · 32 tokens per session scan A 1dccd73e6558
prodx-avaliar is a command published in the GitHub repository bittencourtthulio/expxdev (3 stars, last pushed 1mo ago), licensed MIT. It adds 32 tokens to every session and 621 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-11-07.
prodx-avaliar compared with commands from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| prodx-avaliarbittencourtthulio/expxdev | Command | 3 | 1mo ago | MIT | A | 32 | Claude Code |
| memoryopensolon/soloncode | Command | 198 | 5d ago | MIT | A | 18 | Claude Code |
| dayreportFakeHank/ai-report-to-me | Command | 9 | 8mo ago | MIT | A | 0 | Claude Code |
| qtreportFakeHank/ai-report-to-me | Command | 9 | 8mo ago | MIT | A | 0 | Claude Code |
| modernize-transformfuxicodex/Fuxi | Command | 3.2k | 17d ago | none | A | 17 | Claude Code |
| searchcoltonbearden/carrel | Command | 2 | 3d ago | MIT | A | 29 | Claude Code |
| doctorfatihkan/badi | Command | 7 | 1mo ago | MIT | A | 0 | Claude Code |