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 Orkas-AI/Orkas-Awesome-AgentSkills --skill product-devgit clone --depth 1 https://github.com/Orkas-AI/Orkas-Awesome-AgentSkillsWrote 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/orkas-ai/orkas-awesome-agentskills/product-dev)<a href="https://agentmods.dev/skills/orkas-ai/orkas-awesome-agentskills/product-dev"><img src="https://agentmods.dev/badge/skills/orkas-ai/orkas-awesome-agentskills/product-dev/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/orkas-ai/orkas-awesome-agentskills/product-dev"><img src="https://agentmods.dev/badge/skills/orkas-ai/orkas-awesome-agentskills/product-dev.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 | $0.00108 | $0.01452 |
| Opus 5 | $0.00054 | $0.00726 |
| Sonnet 5 | $0.00022 | $0.00290 |
| Haiku 4.5 | $0.00011 | $0.00145 |
Grade A, and why
product-dev 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 9d 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
Product Dev
用于把“要做什么”拆成“研发如何决策、交付和实现”，并在需要时记录技术方案、总结 spike/POC、指导编码、调试和完成验证。这个 skill 不创建工单、不操作外部项目管理系统、不调度真实多 agent。
何时使用
- 已有 PRD、产品范围、技术约束、设计/测试输入或团队上下文。
- 需要把功能拆成阶段、角色、任务、依赖、交付物、实现要点和验证标准。
- 需要在研发方案设计时记录技术选型、ADR、spike/POC 结论、方案取舍和后果。
- 项目涉及前端、后端、数据、测试、运维、内容等多个研发工作域。
- 需要在编码前明确研发路径、文件/模块影响范围、测试方式和协作边界。
- 需要实际实现代码、修复 bug、按 TDD 推进、调试失败测试或完成前做新鲜验证。
不用于极小个人任务，也不用于替代项目管理系统。
执行方式
- 盘点输入：目标、范围、已有材料、技术约束、角色分工、时间要求、风险。
- 判断是否需要技术决策记录：存在关键技术选型、架构边界、不可逆方案、POC/spike 结论时，先使用 ADR、technical spike 或 decision review reference。
- 识别工作域：前端、后端、数据、测试、运维、内容等；只保留真实需要的研发角色。UI 界面实现可作为独立输入或并行工作流处理，不在这里展开视觉方案。
- 拆阶段：每个阶段交付一个清晰里程碑，避免人为制造过多阶段。
- 拆任务：每个任务必须有 Objective、Output、Validation、Guidance、Dependencies、Steps。
- 建依赖：只写直接依赖；可并行任务明确分组。
- 标人工确认点：范围确认、关键技术选择、设计确认、上线验收、外部凭据或发布动作。
- 补实现要点：对需要编码的任务，列出可能影响的文件/模块、接口、数据、测试方式和预期验证。
- 执行编码任务：需要实际改代码时，按实现、调试和完成验证 references 执行。
- 收尾交接：说明哪些任务已实现，哪些还缺输入或验证。
路由
| 场景 | 读取 |
|---|---|
| 技术选型、架构方案、ADR、关键决策记录 | references/architecture-decision.md |
| 技术预研、POC、spike 总结、可行性结论 | references/technical-spike.md |
| 方案上线前追问、隐藏假设、分支取舍审查 | references/decision-review.md |
| 研发拆解、阶段、任务、依赖、角色交接 | references/product-dev-template.md |
| 实际编码、变更实现、TDD、小步提交前检查 | references/implementation.md |
| Bug、失败测试、回归、异常行为 | references/debugging.md |
| 完成前评审、验证、交付说明 | references/review-and-finish.md |
输出格式
做研发计划时使用 references/product-dev-template.md。实际编码、调试或收尾时只读取对应 reference，不一次加载所有材料。
边界
- 不重写完整 PRD，只摘要影响研发拆解的目标、范围和约束。
- 不把 ADR 或 spike 写成冗长架构论文；只记录决策边界、证据、取舍、后果和下一步。
- 可以写文件/模块级实现要点和测试建议；实际改代码后，必须报告真实执行过的验证。
- 只有在用户任务明确要求实现或修复时才修改项目文件；不自动创建 issue、提交代码或操作外部系统。
- 不把所有传递依赖都写进依赖图，只标直接阻塞关系。
- 信息不足时，输出假设和待确认项，不编造技术事实、团队角色或时间承诺。
质量检查
- 阶段是否对应真实交付里程碑。
- 任务是否有清晰 Objective、Output、Validation、Guidance、Dependencies、Steps。
- 依赖是否只包含直接依赖。
- 可并行任务是否被明确标出。
- 验证标准是否任务级可判定。
- 人工确认点是否清楚。
- 关键技术选择是否记录了备选方案、证据、取舍和后果。
- 交接说明是否能支持下一步编码执行。
- 代码实现是否遵守现有项目模式。
- 完成声明是否有本轮新鲜验证证据。
What ships with it
7 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.
- 9d ago First seen · 74 lines · 0 tokens per session scan A a300a92d7d37
product-dev is a skill published in the GitHub repository Orkas-AI/Orkas-Awesome-AgentSkills (13 stars, last pushed 2mo ago), licensed MIT. It adds 108 tokens to every session and 1,452 once invoked, about $0.0005 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 2026-09-03.
Other skills, from other repositories
viral-tech-reel-editor
End-to-end viral tech reel production for Instagram Reels and TikTok using 2026 trend grammar — retention-first pacing, punch-ins, 3D cinematic AI-generated shots, motion design graphics, proof b-roll, trending SFX, karaoke captions (Georgian/English), safe-zone layout, and QA-gated 1080x1920 export. This skill should…
implementing-aws-config-rules-for-compliance
Implementing AWS Config rules for continuous compliance monitoring of AWS resources, deploying managed and custom rules aligned to CIS and PCI DSS frameworks, configuring automatic remediation with SSM Automation, and aggregating compliance data across accounts.
browser-edge-cases
SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.
Vizra ADK Memory System
Implement persistent memory, session context, and vector memory (RAG) for AI agents.
adaptive-quality
Per-task execution profile selection based on complexity in Balanced quality mode.
auto-qa
QAMESH project QA mesh — plan, run, report, and publish deterministic QA evidence.