translate-to-en

name	translate-to-en
description	Translate/rewrite files into English. Supports PDF, Markdown, plain text, and other formats. Generates an -en.md file in the same directory. Use when user wants to translate documents into English for easier reading.
argument-hint	[file1 file2 ...] or leave empty to be prompted
disable-model-invocation	true

Translate to English

Translate/rewrite files into English to produce a readable English version.

Usage

/translate-to-en                     # No arguments, prompt user for files or directory
/translate-to-en file1.pdf file2.md  # Translate the specified files

Execution Flow

Step 1: Determine the files to translate

If arguments are provided ($ARGUMENTS):

• The arguments are the list of files to translate
• Verify each file exists

If no arguments are provided:

• Use the AskUserQuestion tool to ask the user:
  • Question: "Please provide the file path(s) or a directory path to translate"
  • Options can include: "Enter file path", "Enter directory path", "Cancel"
• If a directory is given, scan it for all non-English files (excluding any that already end in -en.md)

Step 2: Translate files one by one

For each file:

1. Read the source file

   • Use the Read tool to read the file
   • PDF files are automatically parsed into text

2. Generate the translated Markdown file

   • Output path: replace the source file's extension with -en.md
   • Example: /path/to/document.pdf → /path/to/document-en.md
   • Example: /path/to/notes.txt → /path/to/notes-en.md

3. Translation rules (very important):

   The goal is "rewriting", not word-for-word translation:

   • Rewrite using natural English logic, phrasing, and everyday scenarios familiar to English readers
   • Work paragraph by paragraph, preserving the original structure
   • The goal is to make the content easier for English readers to understand

   Preserve original content:

   • Keep all technical terms and domain-specific terminology in their original form (do not over-translate established terms)
   • Keep Chapter titles and Section titles in their original form (they are important anchors)
   • Keep names of people and places in their original form
   • Keep code, formulas, and variable names in their original form

   About examples and scenarios — localize, don't transliterate:

   • When the source contains culturally Chinese examples, replace them with equivalents that feel native to a North American (US) English reader. The goal is for the translated text to read as if it were originally written for a US audience, not as a translation of Chinese content.
   • Personal names: replace Chinese names (张三 / 李四 / 王五, etc.) with common English names (e.g., John Smith, Jane Doe, Michael Johnson, Emily Davis).
   • Company names: replace Chinese companies (阿里巴巴, 腾讯, 字节跳动, 美团, 华为, etc.) with comparable US/North American companies (e.g., Amazon, Google, Meta, Uber, Microsoft) — pick one that matches the role the company plays in the example (e-commerce giant, social platform, ride-hailing, etc.).
   • Software / products / services: replace China-specific apps and services (微信, 支付宝, 淘宝, 滴滴, 钉钉, 高德地图, 百度, etc.) with their Western counterparts (WhatsApp/iMessage, Apple Pay/Venmo, Amazon, Uber/Lyft, Slack, Google Maps, Google, etc.) — again, pick by functional role.
   • Everyday-life scenarios: rewrite scenarios rooted in Chinese daily life (春节抢红包, 双十一, 外卖小哥, 共享单车 in a Chinese city, 国庆假期, etc.) into culturally equivalent North American scenarios (Black Friday / Cyber Monday, holiday tipping, DoorDash delivery driver, Citi Bike in NYC, Thanksgiving weekend travel, etc.).
   • Places & geography: swap Chinese cities/regions (北京, 上海, 深圳, 杭州) for US cities that play a comparable role (New York, San Francisco, Seattle, Austin) when the geographic detail is illustrative rather than essential.
   • Money, units, formats: convert RMB amounts to roughly equivalent USD amounts (round to natural figures, don't compute exact FX), and use US-style date/number/address formats.
   • Exception — preserve the original when:
     • The example is a real, named case study about a specific Chinese company/person/event (e.g., "阿里巴巴 2014 IPO") — keep the real entity, just translate the description.
     • The example's point depends on Chinese specifics (e.g., explaining how 支付宝's QR-code payment flow works) — keep it, and add a brief parenthetical for context if helpful.
     • Technical identifiers, dataset values, code samples, or anything where changing the literal would break the meaning.

   Non-text content handling:

   • Images: use [Image: description] as a placeholder
   • Figures/diagrams: use [Figure: description] as a placeholder
   • Tables: rebuild as Markdown tables when possible; use a placeholder for complex tables

4. Output format:

   # [Original title]
   
   [Translated English content...]
   
   ## [Original Section title]
   
   [Translated English content...]
   
   [Image: original image description]
   
   [Translated English content continues...]
   

Step 3: Report completion

After translation, report:

• The list of successfully translated files and their output paths
• For any failures, explain the reason

Example

User input: /translate-to-en /path/to/lecture1.pdf /path/to/notes.md

Execution:

1. Read lecture1.pdf
2. Generate /path/to/lecture1-en.md (English translation)
3. Read notes.md
4. Generate /path/to/notes-en.md (English translation)
5. Report completion

---

Begin execution: $ARGUMENTS