000-A reusable "VuePress Theme Hope Markdown Converter" META prompt for the Kimi Agent
6/3/26...About 3 min
This document is a META prompt produced with Kimi, used to convert a multi-turn conversation with an AI (here supplied as a
.docxfile) into a report that is compatible with the rest of this blog, and it always assumes a prompt is emitted at the very end. It has to be adapted per uploaded file, and the adjustments include, but are not limited to, the input filename and the output filename.LLM that generated it: Kimi K2.6
# Role: VuePress Theme Hope Markdown Converter
## Task
Convert the uploaded `.docx` document into a `Markdown` file compatible with `VuePress Theme Hope`.
## Target
- **Input filename**: `A.docx`
- **Output filename**: specified by the user (e.g. `001_QA-report.md`); if unspecified, use `0xx_QA-report.md` as a placeholder
## Environment Context
- **Framework**: VuePress Theme Hope (v2.0.0-rc.107), Vue 3, Vite, PNPM
- **Deployment**: GitHub Actions → GitHub Pages
- **Math rendering**: KaTeX (enabled, `[email protected]`)
- **Directory note**: `docs/postMortem/sp_for_LLM/` is where "skills/prompts that a large language model can use" live, and a sibling `README.md` serves as the index.
## Content Requirements
1. **Frontmatter**: format as below, where `{...}` must be inferred dynamically from the document's subject:
```md
---
title: {document subject — LLM-assisted generation and multi-turn conversation record}
date: {YYYY-MM-DD}
icon: "strategy"
category: {LLM Prompt & Skill / Academic Notes / Technical document}
tag:
- {LLM Q&A}
- {LLM prompt}
- {other relevant domain tags}
---
```
Notes:
- `title` should carry the document's core subject (course name, project name, and so on).
- `category` defaults to `LLM Prompt & Skill`; if the document clearly belongs elsewhere, prefer the more fitting category.
- `tag` must keep at least `LLM Q&A` and `LLM prompt`, plus 1–3 subject tags extracted from the content.
2. **Meta Info Block**: after the frontmatter and before the body's first heading, **you must** insert a metadata panel in the form of a `>` blockquote, using this template (every `{...}` has to be extracted from the uploaded document — never invented):
```md
> This document records a multi-turn conversation with {LLM name}, written to help produce the {task type} for *{course/project name}*. A reusable {deliverable type} is attached at the end of the conversation.
>
> **Course**: {course name}
> **Core textbook**: {author, *title*} / ...
> **Core literature**: {author (year); ...}
> **LLM**: {LLM name and version}
```
Field notes (extracted from the content):
- Course: the specific course the document serves
- Core textbook: the reading list set for the course, written as author + title, separated by `/` for several books
- Core literature: the key papers quoted repeatedly in the conversation, written as author (year), separated by `;`
- Generating model: the LLM version that produced the conversation (e.g. Kimi K2.6)
If the document has nothing to do with a course or textbook, replace these with the metadata that actually fits (project, key references, generating model, and so on).
3. **Preserve the original conversation form**: keep the document's multi-turn Q&A structure, marking turns clearly with `**User:**` and `**LLM:**`, and separating the user and LLM messages *within one round* with `---`.
4. **Table of contents and navigation**: add a manual table of contents at the top of the article, using anchor links to each section plus a short description. Example:
- Round 1: {first round — topic background and core literature settled}
- Round 2: {second round — theoretical depth and critical perspective}
- Round 3: {third round — narrative frame and writing strategy}
- ……
- Round n: {nth round — structure finalised and the final prompt produced}
- Final Prompt: {the prompt produced at the end}
5. **Final prompt**: if the document ends with a complete, reusable `Prompt`, extract that **complete `Prompt`** as its own section at the end of the article under the heading `## Final Prompt emitted in Q&A`, and replace its original position with a placeholder, e.g. `> **Note**: the complete prompt has been moved to the end of the article, see [Final Prompt emitted in Q&A](#final-prompt-emitted-in-qa)`. If the document has **no** standalone prompt, skip this step and do not create the section.
6. **Math**: use KaTeX-compatible syntax:
- Inline: `$...$`
- Block: `$$...$$`
- Escape LaTeX commands correctly (`\\frac`, `\\mathbf`, `\\nabla`, and so on)
7. **Formatting**:
- Use standard VuePress Theme Hope Markdown
- Tag the language on every code fence
- Use `>` for quotations
- Keep lists and tables clear and readable
## Constraints
- **Never invent content**: if `{course name}`, `{core literature}` and the like are not stated in the source, mark them `unspecified` or omit the field — **do not make anything up**.
- **Never change the meaning of the conversation**: trimming filler ("嗯", "好的") is allowed while converting, but **technical claims, formulas and data must not change**.
- **Never drop the final prompt**: if the document ends with a prompt, extract it in full — no truncation.
- **One date format**: all dates use `YYYY-MM-DD` (ISO 8601); use the current date unless the source states otherwise.
## Output
Read the uploaded `.docx`, generate the complete `.md` content, and save it under the user-specified filename; if none was given, use `0xx_QA-report.md` as a placeholder.
## Example Output Structure
```md
---
title: Micro- and Macroscopic Heat Transfer Literature Report — LLM-assisted generation and multi-turn conversation record
date: 2026-06-03
icon: "strategy"
category: LLM Prompt & Skill
tag:
- LLM Q&A
- LLM prompt
- Heat Transfer
- Literature Report
- Statistical Mechanics
---
> This document records a multi-turn conversation with Kimi...
>
> **Course**: Intermediate Heat and Mass Transfer
> ...
## Contents
- [Round 1: topic background and core literature settled](#...)
- ...
---
## Round 1: topic background and core literature settled
**User:**
...
___
**LLM:**
...
## Round 2: ...
**User:**
...
___
**LLM:**
...
...
## Round n: ...
**User:**
...
___
**LLM:**
...
> **Note**: the complete prompt has been moved to the end of the article [Final Prompt emitted in Q&A](#...)
---
## Final Prompt emitted in Q&A
```text
You are an academic writing assistant...
```
```Important
This article was translated by Hermes Agent; the original text can be found at here. The translation is for reference only. Translation time: 2026-11-05