Introduction¶
In the plugin ecosystem of DeepSeek Harness (DSH), most plugins focus on logic, code, or data processing. However, in business writing scenarios, a common problem is that the grammar is correct, but the “register” is wrong. For non-native writers, especially native Chinese speakers, common patterns include excessive enthusiasm, excessive apologies, excessive preamble, or being too casual in important situations. These issues are invisible to grammar-checking tools, but they can significantly reduce communication efficiency. This plugin aims to correct the register so that the expression of non-native writers better aligns with business norms.
Plugin Introduction¶
plain-business-english is a plugin focused on calibrating the register of business English. It is maintained by ChenneyZhuang and open-sourced under the MIT license. It is a standalone SKILL.md file with no external dependencies, designed to correct tone issues by non-native writers in business contexts while preserving the author’s personal voice.
Core Features¶
The core logic of the plugin follows a single rule: State facts at their size (state things at the actual size of the facts).
Specific capabilities include:
* Factual scale alignment: Match the energy of a sentence to the size of the fact. Making small matters sound large appears unreliable, while describing important matters calmly is more credible.
* Structured expression: State the action or offer first, then the facts; make requests directly and politely; put the conclusion (Bottom Line) first.
* De-inflation: Remove enthusiasm inflation, apology inflation, certainty inflation, and redundant preambles.
* Precise warmth: Convey warmth through precision rather than jokes; use a hedge only when necessary, and do not overuse them.
Installation and Enabling¶
Run the following command in the terminal to install the plugin into a local directory:
git clone https://github.com/ChenneyZhuang/plain-business-english ~/.claude/skills/plain-business-english
After installation, a DSH process can invoke this skill when handling relevant tasks.
Typical Usage¶
The plugin adjusts tone through comparison and rewriting. The following are typical correction examples based on the README:
* Enthusiasm inflation correction:
* Original sentence: I am genuinely excited to announce our new product.
* After adjustment: State the action or offer first, then the facts.
* Apology inflation correction:
* Original sentence: Sorry to bother you, I just wanted to quickly ask for the file.
* After adjustment: Make a direct, polite request.
* Certainty inflation correction:
* Original sentence: This will definitely solve your problem.
* After adjustment: State the function first, then provide evidence.
* Overly long preamble correction:
* Original sentence: As you know, we have been working on this for months… (the key point is buried in the third paragraph)
* After adjustment: Make the request in the first sentence.
* Inappropriate joke correction:
* Original sentence: Hi boss, just wanted to check in on that thing… (joking when emailing a senior contact)
* After adjustment: Convey warmth through precision, not through jokes.
* Excessive hedging correction:
* Original sentence: maybe we could possibly look into this…
* After adjustment: Use one hedge only when necessary; do not overuse it.
Applicable Scenarios and Notes¶
This plugin is suitable for English business writing by native Chinese speakers, especially when an agent needs to generate or polish English emails or reports.
Note:
* Register norms vary by industry and company culture. The plugin provides defaults, not absolute rules.
* The plugin does not fix weak arguments; it only grounds honest content with its true weight.
* The plugin does not provide native-like fluency; it only calibrates tone.
Conclusion¶
plain-business-english addresses common register issues in non-native business writing through a single rule—stating facts at their actual size. It does not change the content; it only adjusts the measure of expression. It is suitable for scenarios that require high-quality English output. For more details, please refer to the GitHub repository.