# Volunteer Task Extractor Prompt

> Convert volunteer meeting transcripts into structured 'Who Does What By When' tables. Flags missing assignees or deadlines for human review.

## Install

Copy the content below into your project:

# Volunteer Task Extractor Prompt

Convert volunteer meeting transcripts into structured 'Who Does What By When' tables. Flags missing assignees or deadlines for human review.

### Start here
1. **Copy the Prompt**: Copy the complete prompt text provided in the appendix below.
2. **Prepare Input**: Paste your raw transcript of a volunteer coordination call or meeting into the designated area. You may also add optional context notes if specific terms are ambiguous.
3. **Paste and Run**: Open any standard AI chat interface that accepts text input. Paste the full prompt (including your transcript) into the chat box and send it.
4. **Check Output**: The AI will return a Markdown table. Verify that:
   - Action items are extracted, not general chatter.
   - Missing assignees are marked `[MISSING ASSIGNEE]`.
   - Missing deadlines are marked `[MISSING DEADLINE]`.
   - Source quotes accurately reflect the transcript.

### Introduction
This original TokRepo prompt is designed for office and volunteer coordination scenarios. It transforms unstructured, messy conversation logs into clear, actionable task lists. Unlike generic summarization tools, this prompt strictly adheres to explicit commitments, preventing the AI from hallucinating responsibilities or dates. It is ideal for community organizers, event coordinators, and team leads who need to quickly convert verbal agreements into tracked tasks without manual data entry errors.

### Prerequisites and Permissions
- **Access**: You need access to an AI chat interface capable of processing text inputs and generating Markdown tables.
- **Data**: You must have a raw text transcript of a meeting or call. No special API keys or software installations are required.
- **Privacy**: Do not paste sensitive personal information (PII) or confidential data into public AI interfaces unless your organization's policy permits it. Anonymize names if necessary before pasting.

### Limitations
- **Runtime Not Tested**: This guide is based on a static review of the prompt logic. Runtime verification is NOT performed by this worker. Actual output quality depends on the underlying AI model's capabilities.
- **No External Knowledge**: The prompt explicitly forbids the AI from using external knowledge to guess dates. If a date is vague (e.g., "next week"), it will be flagged rather than converted to a specific calendar date.
- **Language Consistency**: While the input can be in any language, the output table headers and flags are fixed in English for consistency.

### FAQ
**Q: What if the transcript does not mention who is responsible for a task?**
A: The prompt is constrained to leave the assignee field blank or mark it as `[MISSING ASSIGNEE]`. It will not invent a name. You must follow up with your team to assign these tasks manually.

**Q: Can this prompt handle multiple speakers?**
A: Yes, the prompt is designed to scan for phrases like "I will..." or "Can someone..." to identify action items regardless of the number of speakers. However, if the speaker identity is unclear in the text, it may default to "Speaker" or leave it blank.

### Attribution
Source: TokRepo Original Prompt (CC-BY-4.0). Reference: ChatGPT release notes. 
*Note: source reviewed; runtime not tested.*

## Complete reusable prompt

# Role: Volunteer Coordination Task Extractor

## Objective
Analyze the provided raw transcript of a volunteer coordination call or meeting. Extract specific action items and structure them into a clear "Who Does What By When" table. 

**Critical Constraint:** You must NOT invent assignees, dates, or deadlines. If a task is mentioned but the responsible person or deadline is not explicitly stated in the text, you MUST leave those cells blank or mark them as `[MISSING]` to flag them for human follow-up.

## Input Data
- **Transcript Text**: A raw, potentially messy text record of a conversation involving multiple speakers.
- **Context Notes** (Optional): Any additional context about the event or project that helps clarify ambiguous terms.

## Step-by-Step Instructions

1. **Identify Action Items**: Scan the transcript for statements that imply a future action, commitment, or task. Look for phrases like:
   - "I will..."
   - "Can someone..."
   - "We need to..."
   - "Let's make sure..."
   - "It's decided that..."

2. **Extract Entities**: For each identified action item, determine:
   - **Task Description**: A concise summary of the action required (max 15 words).
   - **Assignee**: The specific person or role named as responsible. If the speaker says "I'll do it," identify the speaker if possible, otherwise use "Speaker". If no one claims it, leave blank.
   - **Deadline/Date**: The specific date or time mentioned. If vague (e.g., "soon," "by next week"), transcribe the vague phrase but flag it. If no time is mentioned, leave blank.
   - **Status**: Mark as "Committed" (if someone accepted) or "Open/Volunteer Needed" (if no one accepted).

3. **Handle Ambiguity & Missing Info**:
   - If a task is assigned to a group (e.g., "the marketing team"), list the group name. Do not guess individual names.
   - If a deadline is implied but not stated (e.g., "before the event starts"), write `[IMPLICIT: Before Event]` rather than guessing a date.
   - If an assignee is missing, write `[MISSING ASSIGNEE]`.
   - If a deadline is missing, write `[MISSING DEADLINE]`.

4. **Format Output**: Present the result as a Markdown table with the following columns:
   - `ID`: Sequential number (1, 2, 3...)
   - `Task Summary`: The extracted action.
   - `Assignee`: Who is doing it.
   - `Deadline`: When it is due.
   - `Source Quote`: A short excerpt from the transcript proving the extraction.
   - `Human Review Flag`: "Yes" if Assignee or Deadline is missing/implicit; "No" if both are explicit.

## Output Format Example

| ID | Task Summary | Assignee | Deadline | Source Quote | Human Review Flag |
|----|--------------|----------|----------|--------------|-------------------|
| 1 | Print banners for entrance | Sarah | Oct 12 | "Sarah: I can print the banners by Tuesday." | No |
| 2 | Contact local press | [MISSING ASSIGNEE] | [MISSING DEADLINE] | "We really need to get the press involved." | Yes |
| 3 | Set up sound system | Tech Team | [IMPLICIT: Before Event] | "The tech team needs to handle sound." | Yes |

## Constraints & Boundaries
- **Do Not Summarize the Meeting**: Only extract actionable tasks. Ignore general chatter, greetings, or unrelated anecdotes.
- **No External Knowledge**: Do not infer deadlines based on real-world events unless explicitly stated in the text.
- **Strict Fidelity**: If the transcript is unclear, prefer leaving the field blank over guessing.
- **Language**: Process the input in its original language, but output the table headers and standard flags (`[MISSING...]`) in English for consistency.

## User Input Template

Please provide the transcript below:

```
[PASTE TRANSCRIPT HERE]
```

(Optional) Context Notes:
```
[PASTE CONTEXT HERE]
```

## Final Check
Before finalizing, verify:
1. Are all rows actual action items?
2. Are all missing assignees marked `[MISSING ASSIGNEE]`?
3. Are all missing deadlines marked `[MISSING DEADLINE]`?
4. Is the source quote accurate?

Proceed with extraction.

## References and reuse

- [ChatGPT release notes](https://help.openai.com/en/articles/6825453-chatgpt-release-notes)

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

---

<!-- ZH -->

# 志愿者任务提取器提示词

将志愿者会议记录转换为结构化的“谁在何时做什么”表格。标记缺失的负责人或截止日期以供人工审核。

### 开始使用
1. **复制提示词**：复制附录中提供的完整提示词文本。
2. **准备输入**：将志愿者协调会议或通话的原始记录粘贴到指定区域。如果特定术语存在歧义，您也可以添加可选的背景说明。
3. **粘贴并运行**：打开任何接受文本输入的常规 AI 聊天界面。将完整的提示词（包括您的会议记录）粘贴到聊天框中并发送。
4. **检查结果**：AI 将返回一个 Markdown 表格。请验证：
   - 提取的是行动项，而非一般闲聊。
   - 缺失的负责人标记为 `[MISSING ASSIGNEE]`。
   - 缺失的截止日期标记为 `[MISSING DEADLINE]`。
   - 来源引用准确反映了会议记录内容。

### 介绍
此 TokRepo 原创提示词专为办公和志愿者协调场景设计。它将非结构化、杂乱的对话日志转化为清晰、可操作的任务列表。与通用的摘要工具不同，此提示词严格遵循明确的承诺，防止 AI 臆造责任或日期。它非常适合社区组织者、活动协调人和团队领导，能够快速将口头协议转化为可追踪的任务，避免手动数据录入错误。

### 前置条件与权限
- **访问权限**：您需要能够处理文本输入并生成 Markdown 表格的 AI 聊天界面。
- **数据**：您必须拥有会议或通话的原始文本记录。不需要特殊的 API 密钥或软件安装。
- **隐私**：除非贵组织政策允许，否则请勿将敏感个人信息 (PII) 或机密数据粘贴到公共 AI 界面中。如有必要，请在粘贴前对姓名进行匿名化处理。

### 局限性
- **未测试运行时**：本指南基于对提示词逻辑的静态审查。此工作者不执行运行时验证。实际输出质量取决于底层 AI 模型的能力。
- **无外部知识**：提示词明确禁止 AI 使用外部知识来猜测日期。如果日期模糊（例如，“下周”），它将被标记，而不会转换为具体的日历日期。
- **语言一致性**：虽然输入可以是任何语言，但输出表头和标志固定为英语以保持一致性。

### 常见问题
**问：如果会议记录中没有提到谁负责某项任务怎么办？**
答：提示词被约束为将负责人字段留空或标记为 `[MISSING ASSIGNEE]`。它不会凭空捏造名字。您必须与团队跟进，手动分配这些任务。

**问：此提示词能处理多个发言人吗？**
答：是的，该提示词旨在扫描“我将...”或“有人可以...”等短语，无论有多少发言人，都能识别行动项。但是，如果文本中的发言人身份不明确，它可能会默认为“Speaker”或将其留空。

### 致谢
来源：TokRepo 原创提示词 (CC-BY-4.0)。参考：ChatGPT 发布说明。
*注意：source reviewed; runtime not tested.*

## 完整可复制提示词

# 角色：志愿者协调任务提取器

## 目标
分析提供的志愿者协调通话或会议的原始记录。提取具体的行动项，并将其结构化为清晰的“谁在何时做什么”表格。

**关键约束：** 您不得臆造负责人、日期或截止日期。如果文中提到了某项任务但未明确说明责任人或截止日期，您必须将这些单元格留空或标记为 `[MISSING]`，以标记供人工跟进。

## 输入数据
- **记录文本**：涉及多个发言者的对话的原始、可能杂乱的文本记录。
- **背景说明**（可选）：有助于澄清歧义术语的任何关于活动或项目的额外背景信息。

## 分步说明

1. **识别行动项**：扫描记录中暗示未来行动、承诺或任务的陈述。寻找如下短语：
   - “我将...”
   - “有人可以...”
   - “我们需要...”
   - “让我们确保...”
   - “决定...”

2. **提取实体**：对于每个已识别的行动项，确定：
   - **任务描述**：所需行动的简明摘要（最多 15 个字）。
   - **负责人**：被指名负责的具体人员或角色。如果发言者说“我来做”，请尽可能识别发言者，否则使用“Speaker”。如果没有人认领，则留空。
   - **截止日期/日期**：提到的具体日期或时间。如果模糊（例如，“很快”、“下周之前”），请转录该模糊短语但进行标记。如果没有提到时间，则留空。
   - **状态**：如果有人接受，标记为“Committed”（已承诺）；如果无人接受，标记为“Open/Volunteer Needed”（开放/需要志愿者）。

3. **处理歧义与缺失信息**：
   - 如果任务分配给一个团体（例如，“营销团队”），请列出团体名称。不要猜测个人姓名。
   - 如果截止日期是隐含但未说明的（例如，“在活动开始之前”），请写 `[IMPLICIT: Before Event]` 而不是猜测一个日期。
   - 如果缺少负责人，请写 `[MISSING ASSIGNEE]`。
   - 如果缺少截止日期，请写 `[MISSING DEADLINE]`。

4. **格式化输出**：将结果呈现为具有以下列的 Markdown 表格：
   - `ID`：序号 (1, 2, 3...)
   - `Task Summary`：提取的行动。
   - `Assignee`：执行人。
   - `Deadline`：到期时间。
   - `Source Quote`：证明提取内容的记录简短摘录。
   - `Human Review Flag`：如果负责人或截止日期缺失/隐含，则为“Yes”；如果两者均明确，则为“No”。

## 输出格式示例

| ID | Task Summary | Assignee | Deadline | Source Quote | Human Review Flag |
|----|--------------|----------|----------|--------------|-------------------|
| 1 | Print banners for entrance | Sarah | Oct 12 | "Sarah: I can print the banners by Tuesday." | No |
| 2 | Contact local press | [MISSING ASSIGNEE] | [MISSING DEADLINE] | "We really need to get the press involved." | Yes |
| 3 | Set up sound system | Tech Team | [IMPLICIT: Before Event] | "The tech team needs to handle sound." | Yes |

## 约束与边界
- **不要总结会议**：仅提取可操作的任务。忽略一般的闲聊、问候或与主题无关的轶事。
- **不使用外部知识**：除非文中明确说明，否则不要根据现实世界事件推断截止日期。
- **严格忠实原文**：如果记录不清楚，宁愿留空字段也不要猜测。

- **语言**：以原始语言处理输入，但为保持一致性，输出表格标题和标准标记（`[MISSING...]`）时使用英语。

## 用户输入模板

请提供以下会议记录：

```
[PASTE TRANSCRIPT HERE]
```

(可选) 背景说明：

```
[PASTE CONTEXT HERE]
```

## 最终检查
在定稿前，请验证：
1. 所有行是否均为实际的行动项？
2. 所有缺失的负责人是否已标记为 `[MISSING ASSIGNEE]`？
3. 所有缺失的截止日期是否已标记为 `[MISSING DEADLINE]`？
4. 来源引用是否准确？

继续执行提取。

## 参考资料与复用

- [ChatGPT release notes](https://help.openai.com/en/articles/6825453-chatgpt-release-notes)

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/volunteer-task-extractor-prompt-e370e233
Author: Prompt Lab