A command that turns downloaded product-requirements PDFs into versioned Markdown files and sends them for review. A PRD is a document describing what a product or feature should do.

Use it to process PDFs placed in the PRD folder, create the Markdown structure, run the specified review, and remove the source PDF after checking the outputs.

It avoids manually copying PDF content into a repository and provides a repeatable path from source document to reviewed, version-controlled text.

prd-sync — Command for Claude Code by caiaffa

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.

Clone the repo

git clone --depth 1 https://github.com/caiaffa/claude-code-ultimate-engineering-system

Made for: Claude Code.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/caiaffa/claude-code-ultimate-engineering-system/prd-sync/github.svg)](https://agentmods.dev/commands/caiaffa/claude-code-ultimate-engineering-system/prd-sync)

Your own site

<a href="https://agentmods.dev/commands/caiaffa/claude-code-ultimate-engineering-system/prd-sync"><img src="https://agentmods.dev/badge/commands/caiaffa/claude-code-ultimate-engineering-system/prd-sync/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.

Your own site · 80×15

<a href="https://agentmods.dev/commands/caiaffa/claude-code-ultimate-engineering-system/prd-sync"><img src="https://agentmods.dev/badge/commands/caiaffa/claude-code-ultimate-engineering-system/prd-sync.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,961 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00029	$0.01961
Opus 5.5	$0.00012	$0.00784
Sonnet 5.5	$0.00006	$0.00392
Haiku 4.5	$0.00003	$0.00196

Measured 16d ago against content hash 777f0d3ccc18, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-01, from the pricing page.

Security

Grade A, and why

prd-sync 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 16d 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.

.claude/commands/prd-sync.md · 138 lines

How it starts

The opening of the file, as written. The whole thing — 138 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/prd-sync — PDF solto → pasta + Markdown versionado → review

Processa PRDs baixados manualmente do Google Drive como PDF (workaround para Docs bloqueados com "ineligible to be used in generative AI contexts"). O usuário dropa o PDF solto na pasta mãe prds/; este comando cria a estrutura, gera o markdown, delega o review ao subagente principal-engineer, versiona, e por fim deleta o PDF fonte.

Vault: ~/code/kovi/staff/claude/vault/prds/ (referido abaixo como prds/).

Convenções INVIOLÁVEIS do vault

Escrita de conteúdo no vault somente pela tool Write (que cria os diretórios-pai sozinha — nunca use mkdir).
Leitura/descoberta no vault: use as tools Read/Glob — NÃO use Bash sandboxed nesta pasta. O overlay do sandbox tem snapshot stale e já clobberou o diretório (apagou subpastas/metadata). Isso vale inclusive para comandos de leitura como ls/cat.
Bash neste vault só com dangerouslyDisableSandbox: true (FS real, sem overlay), e apenas para o que as tools não fazem: shasum -a 256 do PDF e o rm final do PDF fonte (passo 7). O rm só após verificar que os outputs existem e não estão vazios.
O .md é espelho read-only depois de escrito; o review vai SEMPRE no .review.md irmão. Nunca edite o review dentro do .md.
PRDs e reviews em português.

Passo 1 — Descobrir PDFs a processar

Liste PDFs soltos na raiz com a tool Glob (prds/*.pdf) — não com Bash.
Para cada PDF, calcule shasum -a 256 <pdf> via Bash com dangerouslyDisableSandbox: true (caminho explícito).
Leia prds/.sync-state.json com a tool Read. Se o sha256 já consta como processado → pule (idempotência) e logue "sem alteração".
Se não houver nenhum PDF solto novo, reporte isso e encerre.

Passo 2 — Identificar o PRD e decidir new vs update

Leia o PDF com o Read tool (use pages se >10 páginas; fatie em blocos de ≤20 páginas).
Extraia o título do PRD do conteúdo. Normalize para slug: minúsculo, kebab-case, prefixo prd- (ex.: "PRD - Checklist Remoto Recolha Safety" → prd-checklist-remoto-recolha-safety). Use o nome do arquivo só como fallback se não houver título identificável (e marque para revisão manual no log).
Verifique se prds/<slug>/ existe:
- Não existe → PRD novo, v1.
- Existe → leia o <slug>.md atual e compare o conteúdo:
  - Equivalente (sem mudança real) → pule, logue "sem alteração".
  - Mudou → update, próxima versão vN+1.

Read the full file on GitHub · 138 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.

16d ago First seen · 138 lines · 29 tokens per session scan A 777f0d3ccc18

Subscribe to this mod's changes

prd-sync is a command published in the GitHub repository caiaffa/claude-code-ultimate-engineering-system (17 stars, last pushed 16d ago), licensed MIT. It adds 29 tokens to every session and 1,961 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 2026-09-15.

Documents PDF Project management

prd-sync

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

/prd-sync — PDF solto → pasta + Markdown versionado → review

Convenções INVIOLÁVEIS do vault

Passo 1 — Descobrir PDFs a processar

Passo 2 — Identificar o PRD e decidir new vs update

What this file has done since we first saw it

Other commands, from other repositories

kickoff-review

brief-compliance-check

open

docs

cti-report

cheatsheet