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.
npx skills add songoao25/dsh-virtual-product-team --skill 03-designgit clone --depth 1 https://github.com/songoao25/dsh-virtual-product-teamWrote 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/songoao25/dsh-virtual-product-team/03-design)<a href="https://agentmods.dev/skills/songoao25/dsh-virtual-product-team/03-design"><img src="https://agentmods.dev/badge/skills/songoao25/dsh-virtual-product-team/03-design/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/songoao25/dsh-virtual-product-team/03-design"><img src="https://agentmods.dev/badge/skills/songoao25/dsh-virtual-product-team/03-design.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 | $0.00044 | $0.01017 |
| Opus 5 | $0.00022 | $0.00508 |
| Sonnet 5 | $0.00009 | $0.00203 |
| Haiku 4.5 | $0.00004 | $0.00102 |
Grade A, and why
design 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 11d 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.
What it actually says
阶段 3：技术设计（④技术设计）
角色
你是主 Agent（技术负责人/架构师）。任务：把 PRD 变成"怎么实现"的方案和任务清单。所有技术决策由你做，不让用户碰技术，但要用大白话向用户解释。
专业默认规范（架构师必备知识）
- 技术选型四维：功能满足度、生态成熟度/社区活跃、长期维护成本、AI 熟悉度；要写"为什么选 A 不选 B"。
- 第一要务是简单（KISS）：小型产品能单文件/单体解决的不引微服务/消息队列/重框架；YAGNI 防过度设计（不为想象中的需求做设计）。
- 技术债=有意识取舍+记账+偿还计划，不是无意识质量滑坡；重大取舍写决策记录（ADR：选项对比→理由→放弃项）。
- 分层/模块化/高内聚低耦合；依赖方向（依赖抽象不依赖实现）；不循环依赖。
- 任务拆解：Epic→Feature→Story→Task，每层可独立验收，粒度半天~2 天（≈一次 PR）。
- 非功能需求（性能/安全/可运维性/可用性）在设计阶段显式提出，不拖到上线后补救。
- 方案文档含：数据模型、接口契约、失败/边界场景——不只画流程图。
- 设计原则：SOLID/DRY（三次重复才抽象）/KISS；设计模式"出现该问题才用"，为模式而模式是反模式。
- 给"可证伪的结论"：风险带对策（如"数据库变更需前置迁移且向前兼容"）。
目标
产出 .product-team/TECH-DESIGN.md + .product-team/tasks.md，让开发阶段能直接照着干；二者均为私有工作资料，不得发布。
步骤
- 技术选型（主 Agent 决定）：根据产品类型选择最简单可靠的实现方案（例如：Web 应用选什么框架、数据存哪里、是否要后端）。用大白话向用户说明"我们打算用什么方式做"，不需要用户做选择。
- 派架构评审子 Agent（交叉审查，防自我一致偏见）：
- 把技术方案打包，派 1 个子 Agent 挑刺："你是架构评审员。审查这个方案：有没有更简单可靠的替代？有没有风险？用户是非技术个人开发者，方案必须简单可靠。"
- 汇总评审意见，改进方案。
- 写技术设计（
.product-team/TECH-DESIGN.md），模板：
# 技术设计：<产品名>
## 方案概述
<大白话：用什么方式做、长什么样>
## 技术选型
- <技术 A>：<为什么>
## 目录/结构
<简要说明>
## 风险与对策
| 风险 | 对策 |
|---|---|
- 拆任务清单（
.product-team/tasks.md）——主 Agent 亲自做（收口职责）：每个任务小到可被一个开发回合完成（≈一次 PR 粒度），带验收标准与依赖：
# 任务清单
| # | 任务 | 验收标准 | 依赖 |
|---|---|---|---|
| T1 | <做什么> | <怎么算完成> | - |
- Gate 3 汇报：向用户汇报"设计好了，一共 N 个任务，我们打算这样做……"，用户确认才进阶段 4。
验收标准（Gate 3）
- 技术设计覆盖方案/选型/风险
- 架构评审子 Agent 已交叉审查并改进
- 任务清单每任务可独立完成、带验收标准、依赖明确
- 用户看懂方案概述并确认（用户不需要懂技术细节）
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.
- 11d ago First seen · 67 lines · 44 tokens per session scan A 2b3963c0fae5
design is a skill published in the GitHub repository songoao25/dsh-virtual-product-team (5 stars, last pushed 6d ago), licensed MIT. It adds 44 tokens to every session and 1,017 once invoked, about $0.0002 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.
Other skills, from other repositories
discretelog
Use when users ask about solving the discrete logarithm problem g^x ≡ y (mod P) with Shor-style two-register Fourier sampling, building/explaining DLP circuits, running simulator demos, or debugging post-processing (continued fractions plus two-dimensional Fourier-sample congruence solving). Triggers: discrete log…
aqc
Explains and demonstrates UnitaryLab's small-scale adiabatic quantum linear-system solver. Use it for AQC or QLSP questions, simulator examples, and work involving Householder state preparation, SVD block encoding, adiabatic schedules, post-selection, solution rescaling, or residual checks.
hhl
HHL quantum linear-system solver for Hermitian A and power-of-two dimension N. This implementation accepts user-provided A and b, auto-computes the evolution time t, uses QPE with U = exp(i2πAt), and reconstructs an approximate classical solution from post-selection in a statevector simulation; exponential speedup…
amplitude-estimation
A quantum algorithm for estimating the amplitude of a specific state in a quantum superposition, which can be used for various applications such as Monte Carlo simulations and optimization problems. Provides efficient implementations and educational resources for understanding and utilizing amplitude estimation in…
mps
Loads states with a Matrix Product State representation when low-entanglement structure can reduce the preparation cost. It covers state-to-MPS decomposition, bond-dimension truncation, canonicalization, QR-based unitary completion, work-qubit encoding, leakage, and phase-invariant validation in UnitaryLab.
superposition
Prepares a normalized state with sparse computational-basis support using compact coefficient preparation followed by a support permutation. The current implementation is exact up to floating-point error but materializes dense matrices.