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 Chris1Wang3/Idea-on-Trial --skill vague-requirement-unpackergit clone --depth 1 https://github.com/Chris1Wang3/Idea-on-TrialWrote 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/chris1wang3/idea-on-trial/vague-requirement-unpacker)<a href="https://agentmods.dev/skills/chris1wang3/idea-on-trial/vague-requirement-unpacker"><img src="https://agentmods.dev/badge/skills/chris1wang3/idea-on-trial/vague-requirement-unpacker/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/chris1wang3/idea-on-trial/vague-requirement-unpacker"><img src="https://agentmods.dev/badge/skills/chris1wang3/idea-on-trial/vague-requirement-unpacker.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 | $0.00098 | $0.03260 |
| Opus 5 | $0.00049 | $0.01630 |
| Sonnet 5 | $0.00020 | $0.00652 |
| Haiku 4.5 | $0.00010 | $0.00326 |
Grade A, and why
vague-requirement-unpacker 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 — 217 lines — stays where its author put it; the contents beside it link to each section on GitHub.
模糊需求接招器 · Vague Requirement Unpacker
EN From vague ask -> PM alignment pack.
中文 模糊指令 -> 可对齐的产品行动包。
不是 PRD 生成器,也不是需求评审器,而是 PM 面对“信息不完整但必须推进”的冷启动拆解工具。它先帮用户判断该问什么、怎么回、一期怎么收边界、哪些风险不能承诺。
When / 何时用: 老板或业务方只给了一个模糊方向、口头任务、微信截图、会议纪要短句;需要在 30 分钟到 2 小时内拿出能回老板或能开会对齐的材料;PRD 还没开始写,需要先搭骨架。
Not / 不用: 已有完整 PRD -> 用 pm-requirement-review-simulator;纯竞品调研 -> 用 competitive-product-research;替用户拍板做不做;承诺可直接给研发排期的终稿。
老板说:最近新客转化不太好,看看能不能搞点权益激励。
我现在只有这句话,帮我先拆成能跟老板确认的问题和方案骨架。
可直接触发的说法
以下输入都应触发本技能,不要要求用户先补完整 PRD:
- 老板/业务只给了一个模糊方向,帮我拆一下
- 这个需求现在信息不全,先帮我接住
- 我只有一句会议纪要,帮我拆成对齐问题
- 我还没 PRD,先给我需求冷启动骨架
- 帮我判断这个需求该先问什么
- 帮我把这个模糊需求拆成能开会对齐的材料
- 业务说想提升 XX,但没说清楚怎么做
- 一句话需求,先出接招包
典型输入
老板说:最近新客转化不太好,看看能不能搞点权益激励。
我现在只有这句话,帮我先拆成能跟老板确认的问题和方案骨架。
业务说:能不能给新用户做一个进阶引导,别一上来就流失。
我还没拿到完整需求,先帮我拆成对齐问题、核心场景和一期方案骨架。
会议纪要里只有一句:投顾服务要提升触达后的承接效率。
帮我先拆一下,这到底要确认什么、可能做哪些方向、怎么跟业务对齐。
最小可用输入
最低可启动: 老板原话、业务口头方向、会议纪要短句或截图摘要。缺失信息必须标注为「待确认」,不可脑补成事实。
推荐输入: 老板/业务原话、紧急程度、已知目标或意图、目标用户、核心场景、已知约束、明确不做什么、交付档位、输出格式、后续动作。
可兼容输入:
- 只有一句方向:先提炼可能意图和关键缺口,再展示采集清单。
- 只有会议纪要或截图:先提取可见信息,无法确认的内容保持「待确认」。
- 用户说「直接生成」:只跳过业务信息追问,默认建议 L2;不视为跳过输出格式确认。用户明确委托默认格式时才使用 Markdown。
推荐输入和示例见 references/user_templates.md。
交付档位
| 档位 | 适用场景 | 输出重点 |
|---|---|---|
| L1 回话包 | 30 分钟内要先回应 | 意图还原、关键问题、回话话术、风险红旗 |
| L2 对齐包(默认) | 1-2 小时内要开会对齐 | L1 + 场景收敛、能力骨架、MoSCoW、主流程 |
| L3 开写包 | 要开始写 PRD 草稿 | L2 + 异常分支、埋点指标、非功能、评审交接 |
工作流
1) 接收模糊原话
2) 确认信息采集清单(预填 + 待确认 / 可跳过)
3) Layer 0 模糊度判定 -> 选择 L1 / L2 / L3
4) 按五层接招法展开
5) 计算接招就绪度 + 风险红旗
6) 输出报告 + 可选 handoff 到 pm-requirement-review-simulator
执行细节见 references/unpacking-playbook.md。
信息采集清单(第 2 步)
先预填上下文中的已知事实;未确认值统一标记「待确认」,不得以推断替代用户确认。
1️⃣ 老板/业务原话 2️⃣ 紧急程度 3️⃣ 已知目标或可能意图 4️⃣ 目标用户/角色 5️⃣ 核心场景
6️⃣ 已知约束 7️⃣ 明确不做什么 8️⃣ 交付档位(L1 / L2 / L3) 9️⃣ 输出格式(Markdown / HTML) 🔟 后续动作(仅接招包 / 继续评审预演)
采集交互(优先使用可操作表单):
- 必须复用 assets/intake-form.html,不得由 Agent 临时重写表单 UI;该文件是跨 Agent 稳定展示的唯一实现。
- 通过
prefillURL 参数注入 URI 编码 JSON;原话、交付档位、输出格式为提交门槛,输出格式保持未选。紧急程度可预填;后续动作默认“仅接招包”。 - 原话、意图、用户、场景、约束和不做范围允许编辑;档位、格式与后续动作使用单选。文本清单中的推断值必须标记
[推断];表单预填内容在提交前不得视为已确认。 - 宿主支持时内嵌表单并读取结构化回传;否则复制资产并用浏览器打开,用户粘贴结构化参数后继续。回传统一使用
{schema_version:"1.0", skill, action, data};优先window.codex.submitForm,兼容window.openai.sendFollowUpMessage,最后复制 JSON。仅当 HTML 无法使用时,才退回 references/user_templates.md 文本清单。 - 提交按钮使用“确认清单并开始拆解”;确认后执行 Layer 0 判定,再按已选档位与格式生成。格式未确认不得生成;用户明确委托默认时使用 Markdown。
What ships with it
6 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 · 217 lines · 98 tokens per session scan A f5755914f5c1
vague-requirement-unpacker is a skill published in the GitHub repository Chris1Wang3/Idea-on-Trial (4 stars, last pushed 23d ago), licensed MIT. It adds 98 tokens to every session and 3,260 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 2026-08-31.
Other skills, from other repositories
pilot-competitor-intelligence-setup
Deploy a competitive intelligence system with 4 agents for crawling, analysis, tracking, and alerting. Use this skill when: 1. User wants to set up competitor monitoring or market intelligence 2. User is configuring an agent as part of a competitive intelligence setup 3. User asks about web crawling, market analysis…
pm-sim
This skill has been renamed. Install the new version.
cortex-mem-mcp
Persistent memory enhancement for AI agents. Store conversations, search memories with semantic retrieval, and recall context across sessions. Use this skill when you need to remember user preferences, past conversations, project context, or any information that should persist beyond the current session. Provides…
roblox-input
Use when handling Roblox keyboard, mouse, gamepad, touch, motion input, or cross-platform action binding.
roblox-performance
Use when profiling Roblox performance or diagnosing FPS, memory, network, mobile, or hot-path problems.
changelog-detective
Detect what changed in the product by comparing before and after a deploy or date — find new behaviors, new errors, changed user flows, and unexpected side effects that weren't in the release notes. Use after a deploy or when the user suspects something changed but doesn't know what.