Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Sixz-AI/prd-workflownpx agentmods add skills/sixz-ai/prd-workflow/prd_decomposerWrote 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/skills/sixz-ai/prd-workflow/prd_decomposer)<a href="https://agentmods.dev/skills/sixz-ai/prd-workflow/prd_decomposer"><img src="https://agentmods.dev/badge/skills/sixz-ai/prd-workflow/prd_decomposer/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/skills/sixz-ai/prd-workflow/prd_decomposer"><img src="https://agentmods.dev/badge/skills/sixz-ai/prd-workflow/prd_decomposer.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 | $0.00071 | $0.03527 |
| Opus 5.5 | $0.00028 | $0.01411 |
| Sonnet 5.5 | $0.00014 | $0.00705 |
| Haiku 4.5 | $0.00007 | $0.00353 |
Grade A, and why
prd-decomposer 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PRD Decomposer Skill
Overview
This skill transforms complex, lengthy PRDs (Product Requirement Documents) into purely structural documentation (a set of Markdown files) defining optimally sized requirement units (REQs).
Core Focus: A high-quality decomposition focuses on WHAT we are building and the underlying Business Logic, never on HOW to code it. The Agent proactively identifies ambiguous logic, confirms boundaries, and breaks the logic into atomic manageable pieces without dictating implementation methods to downstream agents.
When to Use
- User provides a new PRD or feature requirement document
- The business logic needs to be pressure-tested and clarified before development
- Requirements need to be broken down into atomic business logic requirement units (REQs)
Core Principles
1. Clarify First, Decompose Later Never blindly parse a poorly defined PRD. Always interactively force the user to confirm ambiguous logic, edge cases, and boundaries first.
2. Code Exploration as INPUT, Not OUTPUT When the user provides a codebase reference (e.g., project path), thoroughly explore the code to understand existing functionality, current behavior, and business context. This understanding is your INPUT — it helps you describe the current state accurately, identify affected features, and fill in PRD gaps. However, the OUTPUT must be expressed in pure business language. NEVER include file paths, line numbers, variable names, function names, or architectural patterns in the generated documents. The downstream Dev Agent will explore the code themselves.
3. Focus on Business Logic (What, Not How) Requirement breakdown is strictly about decomposing Business Logic, not dictating implementation. Clearly explain WHAT the feature is and WHAT the business rules are. NEVER tell the downstream AI "how to code it" (e.g., do not suggest using arrays, specific hooks, or CSS frameworks unless strictly mandated by the PRD). Do NOT create REQs that are purely about code refactoring or architectural migration — if it's not a business behavior change, it doesn't belong in the breakdown.
4. Current State → Target State → Delta Every REQ must clearly describe: what exists today (现状), what the product wants (需求), and the precise difference (变更: add/modify/remove). This framing helps the downstream Dev Agent understand the full business context and make informed implementation decisions.
5. File-Based Progressive Disclosure Output the breakdown as a structured directory of files, starting with an outline index.
What ships with it
4 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 1mo ago First seen · 259 lines · 71 tokens per session scan A 719fe9401731
prd-decomposer is a skill published in the GitHub repository Sixz-AI/prd-workflow (3 stars, last pushed 6mo ago), licensed MIT. It adds 71 tokens to every session and 3,527 once invoked, about $0.0003 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 2026-08-31.
Other skills, from other repositories
dev-plan
Transforma um BRIEF (ou ideia já discutida) em PLAN.md atômico — tasks com critério verificável, vertical slices, must-haves observáveis, esforço estimado e pontos de reset de contexto. Sem código no plano. Auto-suficiente para reset — uma nova sessão consegue retomar lendo só o PLAN.md. Use quando o usuário disser…
dev-brainstorm
Grilling estruturado para estressar uma ideia de feature ANTES de planejar. Traduz ideia falada/solta em requisitos, classifica o tamanho (S/M/L), faz uma pergunta por vez sempre com recomendação inline, explora o codebase para resolver dúvidas sem perguntar, captura decisões em BRIEF.md ao vivo e fecha com radar de…
dev-ship
Fechamento de feature — verificação goal-backward antes de declarar pronto. Roda Must-Haves do PLAN.md, executa o demo script, revisa o diff completo procurando bugs e restos (debug logs, TODOs, código morto), passa lente de segurança nos arquivos tocados, escreve SUMMARY.md e arquiva o plano. Use quando o usuário…
council-review
Perform a rigorous Carmack Council code review. Use when explicitly asked to review code, do a "council review", "carmack review", or invoke /council-review. Carmack's philosophy chairs a council of domain experts — Troy Hunt (security), Martin Fowler (refactoring), Kent C. Dodds (frontend), Matteo Collina (Node.js)…
dev-fix
Loop de diagnóstico disciplinado para bugs — sem PLAN.md, direto ao ponto. Constrói feedback loop reproduzível primeiro, ranqueia hipóteses falsificáveis, instrumenta um probe por hipótese, corrige com teste de regressão e limpa os rastros. Tem modo rápido para bug trivial. Use quando o usuário disser "/dev-fix", "tem…
council-plan
Architect a feature with the Carmack Council before writing code. Use when explicitly asked to plan a feature, do a "council plan", "carmack plan", or invoke /council-plan. Carmack's philosophy chairs a council of domain experts — Troy Hunt (security), Martin Fowler (refactoring), Kent C. Dodds (frontend), Matteo…