# Stakeholder Interest Map from Meeting Notes

> Paste meeting notes into this prompt to get a neutral table of every named party's stated position, basis, open questions and evidence strength.

## Install

Copy the content below into your project:

# Stakeholder Interest Map from Meeting Notes

Paste meeting notes into this prompt to get a neutral table of every named party's stated position, basis, open questions and evidence strength.

## Start here

Copy the complete original prompt text (published with its full Chinese translation) into any ordinary AI chat that accepts pasted text. No terminal, account setup or API is required.

**What to paste before the prompt:**
- Meeting name, date, and your role (if known) — fill the `{{...}}` slot.
- The raw notes, transcript excerpt, chat log or email thread.
- Optionally, the upcoming decision or question you care about.

**Check the output before you rely on it:** every name in the table must appear by name in your notes; each quoted phrase must be short and exact; the map must not claim agreement, decisions or a majority unless your notes state it. If the notes name nobody, the correct output is to say so and stop.

## What this prompt does

It reads only the text you paste and builds a stakeholder-interest map: a table of each named person, team, role or organization, with their stated position, the basis given, any unresolved question, and an evidence-strength label (`explicit`, `implied`, `unclear`). Below the table it adds shared ground, live disagreements, unattributed or floating issues, and a list of things you still need to confirm with real people before relying on the map.

It is a preparation aid for your own review, not a record of what was decided.

## Prerequisites and permissions

- A general AI chat that accepts pasted text. Nothing else is needed.
- You must have the right to paste those notes into that tool. Meeting notes and chat logs often contain names, commercial plans or personal remarks; redact anything you are not allowed to share.
- The prompt cannot browse, message anyone, schedule meetings or update any system; it only reads what you paste.

## Limitations

- Quality depends entirely on your notes. Garbled or partial quotes are marked `[partial quote]` and left incomplete.
- Contradictory notes about the same party are shown side by side, marked `conflicting notes`.
- An `implied` or `unclear` label means the model is inferring; treat it as a prompt to verify, not a fact.
- `reason not stated` and `no explicit open question in these notes` are required placeholders instead of guessing.
- It gives no legal, HR, medical or financial advice, and it is not a final record of commitments.

## FAQ

**Can I use it with a rough transcript full of filler words?**
Yes, as long as names are intact. Expect many `reason not stated` entries, since the checklist requires those placeholders instead of guessing.

**It invented a person who was only mentioned in passing — what now?**
Reject that row and edit the notes to remove or flag ambiguous names. The prompt's own check says every party must appear by name in your notes.

## Verification note

Source reviewed; runtime not tested. The supplied evidence includes no executed run, so behaviour across specific chat products is not claimed. The worked example in the original prompt is labelled fictional and is an output-shape illustration, not a test result.

## Source

Original TokRepo prompt, CC BY 4.0. Reference context: [ChatGPT release notes](<https://help.openai.com/en/articles/6825453-chatgpt-release-notes>) (reviewed 2026-10-04). Thanks to the contributors of that reference material; external material keeps its own rights.

## Complete reusable prompt

You help me turn supplied meeting notes into a stakeholder-interest map. Work only from the text I provide. Do not invent positions, alliances, priorities, decisions, or quotes.

## Input I will paste
- Meeting name, date, and my role (if known)
- The raw notes, transcript excerpt, chat log, or email thread
- Optional: upcoming decision or question I care about

## Task
1. Read my notes fully before writing anything.
2. Identify every named person, team, role, or organization that appears. Do not add parties who are not named.
3. For each party, extract:
   - **Stated position**: Their view or preference as expressed in the notes. Quote a short phrase (max 15 words) from the notes and label it as a quote.
   - **Basis or reason**: Why they hold that view, if stated. If not stated, write `reason not stated`.
   - **Unresolved question**: Any open question, objection, condition, or missing information connected to that party. If none appears, write `no explicit open question in these notes`.
   - **Evidence strength**: `explicit` if the notes directly state it, `implied` if it can only be inferred, `unclear` if ambiguous.
4. After the table, add:
   - **Shared ground**: points where two or more parties appear aligned, with the supporting quote or line.
   - **Live disagreements**: points where positions conflict, showing both sides.
   - **Unattributed or floating issues**: concerns in the notes not tied to a named party.
   - **My action list**: what I still need to confirm with real people before relying on this map.

## Output format
Markdown with:
- A one-paragraph `Scope note` stating which source text you used and its apparent completeness.
- A table with columns: Party | Stated position (quoted if possible) | Basis | Unresolved question | Evidence strength.
- The four sections above under clear headings.
- A final `Boundaries` line: "This map reflects only the supplied notes. It is not a record of decisions, commitments, or consensus."

## Uncertainty and missing data
- If the notes do not name any parties, say so and stop.
- If a quote is partial or garbled, mark it `[partial quote]` and do not complete it.
- If two notes contradict each other about the same party, show both and mark `conflicting notes`.
- If I provide an upcoming decision, flag which parties' unresolved questions matter most for it, using only the notes.

## Checks before you finalize
- Every party in the table appears by name in my notes.
- Every quoted phrase is copied exactly and short.
- No claim of agreement, decision, or majority unless the notes state it.
- `reason not stated` and `no explicit open question in these notes` used where appropriate instead of guessing.
- The action list contains only verification steps, not actions you performed.

## Boundaries
- Do not browse, message anyone, schedule meetings, or update any system.
- Do not give legal, HR, medical, or financial advice.
- Do not treat this map as a final record; it is a preparation aid for my review.

## Worked example
Fictional input:
```
Meeting: Q3 launch sync, 12 May
Ana (Marketing): wants the launch pushed to June because the campaign assets are not ready.
Ben (Sales): needs it in May to hit the quarter, says pipeline depends on it.
Priya (Support): asks whether training will happen before either date; no answer given.
```
Illustrative output shape:
| Party | Stated position | Basis | Unresolved question | Evidence strength |
|---|---|---|---|---|
| Ana (Marketing) | "wants the launch pushed to June" | "campaign assets are not ready" | none stated in notes | explicit |
| Ben (Sales) | "needs it in May" | "pipeline depends on it" | none stated in notes | explicit |
| Priya (Support) | no position stated | reason not stated | "whether training will happen before either date; no answer given" | explicit |
Shared ground: none stated. Live disagreement: Ana vs Ben on timing. Floating issue: training timing (Priya). Action list: confirm whether training date is decided; confirm Ana's asset readiness date.

Checks for this example would pass if all three names come from the notes, each quote is short and exact, no consensus is claimed, and the action list contains only follow-up questions. Fail if the output invents a fourth party, upgrades Priya's question into a position, or states that a June launch was agreed.

## Input now
Meeting name, date, role: {{...}}
Upcoming decision or question I care about (optional): {{...}}
Notes: {{paste here}}

## References and reuse

- [ChatGPT release notes](https://help.openai.com/en/articles/6825453-chatgpt-release-notes) · Reviewed 2026-10-04

Original TokRepo prompt · [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/). Reference documents retain their own rights.

---

<!-- ZH -->

# 根据会议记录生成利益相关方关注点地图

把会议记录粘贴进这个提示词，得到一份中立表格：逐个列出被点名相关方的明确立场、依据、未决问题与证据强度。

## 开始使用

把完整的原始提示词文本（发布方会附上完整中文翻译）复制到一个能接受粘贴文本的普通 AI 聊天窗口中。无需终端、账号配置或 API。

**提示词之前要先粘贴的内容：**
- 会议名称、日期，以及你的角色（如果知道）——填入 `{{...}}` 位置。
- 原始记录、转写片段、聊天记录或邮件线程。
- 可选：你关心的即将做出的决定或问题。

**使用前先检查输出：** 表格中的每个名字都必须在你提供的记录里被点名出现；每条引语必须简短且原样照抄；除非记录明确写着，否则不得声称达成一致、作出决定或形成多数意见。如果记录里没有点名任何人，正确做法是说明这一点并停止。

## 这个提示词做什么

它只读取你粘贴的文本，生成一份利益相关方关注点地图：用表格逐个列出被点名的人、团队、角色或组织，包括明确立场、给出的依据、未决问题，以及证据强度标记（`explicit`、`implied`、`unclear`）。表格之后还会补充共同点、实际分歧、未归属或无人认领的问题，以及你仍需向真人核实后才能依赖这份地图的事项清单。

它是供你自己复核的准备工具，不是会议决定的记录。

## 前提与权限

- 一个能粘贴文本的普通 AI 聊天工具，无需其他条件。
- 你必须有权限把这些记录粘贴进该工具。会议记录和聊天记录常含姓名、商业计划或私人评论，不应外传的内容请先删减。
- 提示词无法浏览网页、给任何人发消息、安排会议或更新任何系统；它只读取你粘贴的内容。

## 局限

- 效果完全取决于你的记录。残缺或混乱的引语会被标记为 `[partial quote]`，且不会补全。
- 同一方的记录互相矛盾时，两边都会列出并标记 `conflicting notes`。
- 标记为 `implied` 或 `unclear` 表示模型在推断，应视为待核实的提示，而非事实。
- `reason not stated` 和 `no explicit open question in these notes` 是必须使用的占位符，不能以猜测代替。
- 它不提供法律、人力资源、医疗或财务建议，也不是承诺的最终记录。

## 常见问题

**可以用充满口头语的水货转写稿吗？**
可以，只要人名完整。预计会出现很多 `reason not stated`，因为检查清单要求用这些占位符而不是猜测。

**它凭空加了一个只是被顺带提到的人，怎么办？**
删掉那一行，并在记录中移除或标记含义不明的名字。提示词自带的检查要求：每个相关方都必须在你提供的记录中被点名。

## 核实说明

来源已审阅；未做运行测试。所提供材料中没有实际运行记录，因此不对具体聊天工具中的表现作任何断言。原始提示词中的示例已标注为虚构，它只是输出形式的示意，不是测试结果。

## 来源

TokRepo 原创提示词，CC BY 4.0。参考背景：[ChatGPT release notes](<https://help.openai.com/en/articles/6825453-chatgpt-release-notes>)（审阅于 2026-10-04）。感谢该参考材料的贡献者；外部材料保留其自身权利。

## 完整可复制提示词

你帮我将提供的会议记录转成一份利益相关方关注点地图。只依据我提供的文本进行。不要虚构立场、联盟、优先级、决定或引语。

## 我将粘贴的输入
- 会议名称、日期，以及我的角色（如已知）
- 原始记录、转写片段、聊天记录或邮件线程
- 可选：我关心的即将做出的决定或问题

## 任务
1. 在写任何内容之前先完整阅读我的记录。
2. 识别所有被点名的人、团队、角色或组织。不要添加未被点名的相关方。
3. 对每个相关方，提取：
   - **明确立场**：他们在记录中表达的观感或偏好。从记录中引用一个简短短语（最多 15 个词）并标记为引语。
   - **依据或理由**：他们为何持该观感（如有陈述）。如未陈述，写 `reason not stated`。
   - **未决问题**：与该相关方相关的任何未决问题、异议、条件或缺失信息。如无，写 `no explicit open question in these notes`。
   - **证据强度**：如果记录直接陈述则为 `explicit`，如果只能推断则为 `implied`，如果含义不明则为 `unclear`。
4. 表格之后，补充：
   - **共同点**：两方或多方观感一致的地方，附支持性引语或行。
   - **实际分歧**：立场冲突的地方，展示双方。
   - **未归属或无人认领的问题**：记录中未与任何被点名相关方关联的关切。
   - **我的行动清单**：在依赖这份地图之前，我仍需与真人确认的事项。

## 输出格式
Markdown，包含：
- 一段 `Scope note`，说明你使用了哪段源文本及其表面完整度。
- 一个表格，列为：Party | Stated position (quoted if possible) | Basis | Unresolved question | Evidence strength。
- 上述四个部分，使用清晰标题。
- 末尾一行 `Boundaries`：“This map reflects only the supplied notes. It is not a record of decisions, commitments, or consensus.”

## 不确定性与缺失数据
- 如果记录未点名任何相关方，说明这一点并停止。
- 如果引语残缺或混乱，标记为 `[partial quote]` 且不要补全。
- 如果记录中同一相关方存在互相矛盾之处，两边都展示并标记 `conflicting notes`。
- 如果我提供了即将做出的决定，仅依据记录指出哪些相关方的未决问题对该决定最重要。

## 定稿前的检查
- 表格中的每个相关方都在我的记录中按姓名出现。
- 每条引语都原样照抄且简短。
- 除非记录明确写着，否则不得声称达成一致、作出决定或形成多数意见。
- 在适当处使用 `reason not stated` 和 `no explicit open question in these notes` 而不是猜测。
- 行动清单只包含核实步骤，不包含你执行的动作。

## 界限
- 不要浏览网页、给任何人发消息、安排会议或更新任何系统。
- 不要提供法律、人力资源、医疗或财务建议。
- 不要将此地图视为最终记录；它是供我复核的准备工具。

## 已完成的示例
虚构输入：
```
Meeting: Q3 launch sync, 12 May
Ana (Marketing): wants the launch pushed to June because the campaign assets are not ready.
Ben (Sales): needs it in May to hit the quarter, says pipeline depends on it.
Priya (Support): asks whether training will happen before either date; no answer given.
```
示意性输出形式：
| Party | Stated position | Basis | Unresolved question | Evidence strength |
|---|---|---|---|---|
| Ana (Marketing) | "wants the launch pushed to June" | "campaign assets are not ready" | none stated in notes | explicit |
| Ben (Sales) | "needs it in May" | "pipeline depends on it" | none stated in notes | explicit |
| Priya (Support) | no position stated | reason not stated | "whether training will happen before either date; no answer given" | explicit |
共同点：未陈述。实际分歧：Ana 与 Ben 在时间安排上。无人认领的问题：培训时间（Priya）。行动清单：确认培训日期是否已决定；确认 Ana 的资产就绪日期。

对这个小节而言，只有在以下情况全部满足时检查才算通过：三个名字都来自你的记录，每条引语都简短且原样照抄，没有声称任何一致意见，并且行动清单只包含后续待问的问题。如果输出凭空捏造出第四个相关方、把 Priya 的提问升级为一个立场，或者声称已就 6 月上线达成一致，则判定为不通过。

## 现在输入
会议名称、日期、角色：{{...}}
我关心的即将做出的决定或问题（可选）：{{...}}
记录：{{paste here}}

## 参考资料与复用

- [ChatGPT release notes](https://help.openai.com/en/articles/6825453-chatgpt-release-notes) · Reviewed 2026-10-04

TokRepo 原创提示词 · [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/)。参考资料保留各自原有权利。

---
Source: https://tokrepo.com/en/workflows/stakeholder-interest-map-meeting-notes-50a4b899
Author: Prompt Lab