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 echoyu1025-a11y/ai-product-skills --skill ai-product-proposalgit clone --depth 1 https://github.com/echoyu1025-a11y/ai-product-skillsWrote 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/echoyu1025-a11y/ai-product-skills/ai-product-proposal)<a href="https://agentmods.dev/skills/echoyu1025-a11y/ai-product-skills/ai-product-proposal"><img src="https://agentmods.dev/badge/skills/echoyu1025-a11y/ai-product-skills/ai-product-proposal/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/echoyu1025-a11y/ai-product-skills/ai-product-proposal"><img src="https://agentmods.dev/badge/skills/echoyu1025-a11y/ai-product-skills/ai-product-proposal.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 | $0.00159 | $0.02074 |
| Opus 5 | $0.00079 | $0.01037 |
| Sonnet 5 | $0.00032 | $0.00415 |
| Haiku 4.5 | $0.00016 | $0.00207 |
Grade A, and why
ai-product-proposal 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 12d 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AI 产品方案 SKILL — 从想法到完整方案
这个 SKILL 是什么
把"一个 AI 产品想法"通过结构化问卷推导成产品经理可交付的完整方案:
- 四层骨架文字版:触达 / 场景 / 能力 / 数据 各层有几个模块、每个模块做什么、为什么这么切
- 数据流转:至少 1 条向下沉淀 + 1 条向上回流的清晰描述
- 关键判断:为什么选这个层级粒度、哪些是核心假设、哪些是待验证空白点
- 输出 Markdown 文字方案(不是可视化图)
正向产出,不是逆向拆解。如果用户要拆解一个已有产品,转 product-architecture skill。
跟 product-architecture 的关系
| product-architecture | ai-product-proposal(本 skill) | |
|---|---|---|
| 方向 | 逆向拆解 | 正向设计 |
| 输入 | 已有产品名 | 产品想法 + 问卷回答 |
| 输出 | HTML 可视化 | Markdown 文字方案 |
| 核心问 | "他怎么搭的?" | "我应该怎么搭?" |
两个 skill 共享同一套方法论(四层骨架 / 数据流转 / 粒度判断)。本 skill 不重写方法论,直接调用老 skill 的 references。
触发后的工作流程
第 0 步:确认场景与用途
如果用户给的信息不够,先问 5-7 个核心问题,信息齐了再生成。
问卷模板(按顺序问,可以一次性问完也可以分批):
| # | 问题 | 用途 |
|---|---|---|
| 1 | 产品名 + 一句话定位 | 锁定输出主体 |
| 2 | ToC 还是 ToB? | 决定是否需要 4A 体系/生态层 |
| 3 | 目标用户 + 核心痛点 | 撑起场景层的"为什么用" |
| 4 | 你设想的核心场景有哪些?(3-5 个) | 直接对应场景层模块 |
| 5 | 已知的能力/数据约束?(可选,留空我会推理) | 撑起能力层和数据层的现实性 |
| 6 | 输出用途:① 立项汇报 ② PRD 前期 ③ 面试案例 ④ 其他 | 决定输出形态(完整方案 vs 精简大纲) |
| 7 | 详细程度:精简大纲(快速看判断)/ 完整方案(可直接贴 PRD) | 用户没说就默认完整方案 |
关键判断:
- 如果第 6 题选了面试案例,跳到第 4 步:面试草稿模式。
- 如果用户提供的产品是他不熟悉的(常见于面试题),用最少假设 + 最常见的 AI 产品形态推理,明确标注哪些是假设。
第 1 步:阅读方法论
读同级目录下 ../product-architecture/references/methodology.md,这是四层骨架方法论的核心,跟本 skill 共享。重点掌握:
- 四层骨架定义(触达/场景/能力/数据各层是什么)
- 数据流转的颜色和方向语义(虽然本 skill 输出文字,但语义要对)
- 粒度判断("能描述功能帮用户做什么"就够了)
- 归类判断("砍掉 A 后 B 是否受影响"测试)
- ToB/ToC 差异(ToB 多 4A 体系,AI ToC 可能有生态层)
第 2 步:阅读正向推导示范
读本 skill 的 references/methodology.md(正向推导版本)和 references/examples.md(AI 客服完整方案示范)。
老 skill 的 examples.md 是"逆向拆解后的形态",本 skill 的 examples.md 是"从想法正向写到方案的形态",两者侧重点不同。
第 3 步:六步法正向推导
按顺序推导,输出方案前自查每一步:
- 从用户痛点出发(对应问卷 Q3),确认目标用户和核心痛点
- 穷举设想场景(对应问卷 Q4),如果用户给的少于 3 个,引导补齐
- 归类成场景层模块(3-6 个,用"砍掉 A 影响 B"测试)
- 反推能力层:每个场景需要什么 AI/通用能力?共用的只画一次
- 反推数据层:每个能力消费什么数据?用业务语言不用技术语言
- 设计数据流转:至少 1 条向下沉淀(产品 → 数据)+ 1 条向上回流(数据 → 产品)
第 4 步:生成输出
完整方案场景(立项/PRD):
- 复制
templates/proposal-full.md→ 当前工作目录 - 文件名:
<产品名>-proposal.md - 填充所有占位符,标注待验证假设
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.
- 12d ago First seen · 127 lines · 159 tokens per session scan A 620ee0311fdd
ai-product-proposal is a skill published in the GitHub repository echoyu1025-a11y/ai-product-skills (3 stars, last pushed 2mo ago), licensed MIT. It adds 159 tokens to every session and 2,074 once invoked, about $0.0008 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
shortfilm-prompt
Generate cinematic AI shortfilm prompts (works with Seedance 2.0, Xiaoyunque, Sora, Kling, Jimeng, Veo) using the 5-stage structure from Mx-Shell's Zombie Scavenger. Trigger when the user wants transformation sequences, multi-shot narrative shorts, weapon-charge/combat segments, emotional family/pet/farewell…
guochao-visual-cn
A prompt-writing guide for generating Chinese-inspired illustrations and other visual designs. It distinguishes twelve Chinese artistic styles, such as ink painting, gongbi painting, blue-green landscapes, and Dunhuang murals.
prompt-design
A structured method for designing prompts, the instructions given to an AI system. It covers the situation, the desired output, constraints, context, user input, and replaceable variables.
prompt-evaluation
A structured method for testing and comparing prompts, the instructions given to an AI. It uses defined criteria, test examples, comparisons, scores, and a recommendation.
searchcans-rag-source-curator-zh
A Chinese-language tool for building a small, varied list of web pages and documents that can support answers in a retrieval-augmented generation (RAG) system. RAG systems retrieve source material to ground an AI response.
agent-orchestration-improve-agent
Systematic improvement of existing agents through performance analysis, prompt engineering, and continuous iteration.