Agent skill

Blog Writing Style

by eunomia-bpf in eunomia-bpf/eunomia.dev

Define the finished prose style for English and Chinese eunomia.dev blog posts and Research pages.

MITAuto-check passedWriting & Content

Install Blog Writing Style

skills CLI
$ npx skills add eunomia-bpf/eunomia.dev --skill blog-writing-style -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install eunomia-bpf/eunomia.dev blog-writing-style --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/eunomia-bpf/eunomia.dev.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/blog-writing-style .claude/skills/blog-writing-style && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
blog-writing-style
GitHub stars
236
Token cost
~2.5k tokens
SKILL.md length
1,372 words
Files
1
Skills in repo
18
Repo updated
First seen
Licence
MIT

At a glance

Define the finished prose style for English and Chinese eunomia.dev blog posts and Research pages.

  • Works in 6 steps: a concrete situation or failure; → the incomplete mental model that made it… → the mechanism; → …
  • Reviewing reader comfort
  • SKILL.md covers Human Reader First, Progressive Disclosure, Terms And Abstractions and Reading Experience, plus 7 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Blog Writing Style is an agent skill from eunomia-bpf/eunomia.dev. Define the finished prose style for English and Chinese eunomia.dev blog posts and Research pages. Use when writing or reviewing reader comfort, natural rhythm, technical tone, research readability, and recurring sentence-level AI tells. This skill does not choose the topic or evidence.

Its SKILL.md is about 2.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Writing & Content, covering Blog and article writing and Plain language and style rules. The repository describes itself as: https://github.com/eunomia-bpf homepage, documents and blogs. The licence is MIT.

When your agent uses it

  • Reviewing reader comfort
  • Research readability
  • Recurring sentence-level AI tells

Example prompts

  • “/blog-writing-style”

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. a concrete situation or failure;
  2. the incomplete mental model that made it possible;
  3. the mechanism;
  4. the design or decision that changes;
  5. evidence, implementation detail, and alternatives;
  6. limits and falsification.

What it can do on your machine

Read from SKILL.md and the folder at commit 4cc0b4f. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Blog Writing Style loads about 2.5k tokens when it runs. Until then it costs about 77 tokens; SKILL.md has 1,372 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~77
When it runs · the whole SKILL.md, loaded when a task matches
~2.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from eunomia-bpf/eunomia.dev at commit 4cc0b4f, republished under its MIT licence (© eunomia-bpf). 1,372 words, ~2,483 tokens.

Download SKILL.mdSave it as .claude/skills/blog-writing-style/SKILL.md (or your agent's skills folder).
name
blog-writing-style
description
Define the finished prose style for English and Chinese eunomia.dev blog posts and Research pages. Use when writing or reviewing reader comfort, natural rhythm, technical tone, research readability, and recurring sentence-level AI tells. This skill does not choose the topic or evidence.

Blog Writing Style

Judge the finished prose by one outcome: a technically qualified human reader can read it comfortably once, understand the argument, and explain the main decision without reconstructing missing premises.

This is a style reference, not a source or workflow specification. Examples identify failure patterns, not banned words or sentence-shape quotas.

Human Reader First

  • Write for a domain-aware engineer, researcher, or maintainer who has not read the source corpus.
  • Establish the concrete situation, actor, state change, and consequence before introducing a taxonomy, formal model, or coined abstraction.
  • Make the title and description understandable without prior knowledge of a paper, project, or invented term.
  • Within the opening screen, tell the reader what problem exists, why it matters, and what the article argues.
  • A report may be deep without sounding like a paper abstract. Academic quality comes from evidence and reasoning, not compressed terminology.
  • Organize sections around questions the reader needs answered, not around the order in which sources were discovered.
  • A long article should remain selective. Include detail that changes the reader's model or decision; remove detail that only proves the author read many sources.

Progressive Disclosure

Use this order when it fits the subject:

  1. a concrete situation or failure;
  2. the incomplete mental model that made it possible;
  3. the mechanism;
  4. the design or decision that changes;
  5. evidence, implementation detail, and alternatives;
  6. limits and falsification.

Do not front-load a dense comparison table, a list of runtime names, equations, or a newly named property. Introduce these only after the reader knows what they explain.

When a formal definition is useful, first explain the same idea in ordinary language and with an example. Define every symbol locally. If removing the notation does not reduce precision, remove it.

Terms And Abstractions

  • Use the fewest terms needed to carry the argument.
  • Define a term before relying on it. A coined term should name a pattern the reader has already seen, not create the appearance of novelty.
  • Prefer one stable term to several near-synonyms.
  • Do not make adjacent headings stacks of English nouns or unexplained abstractions.
  • Avoid paragraphs containing several new terms at once. Introduce one idea, show its consequence, then continue.
  • After a dense technical section, give the reader a plain-language implication before moving on.
  • Source names support the argument; they do not substitute for it. Do not write a paper-by-paper parade.

Reading Experience

  • Write clear, natural, technically serious prose rather than notes, an abstract, a benchmark ledger, or promotion copy.
  • Give enough local context for the next sentence without forcing the reader to infer a necessary premise or a change of subject.
  • Let each thought lead naturally to the next. Use transitions that express the actual causal, conditional, or contrasting relationship.
  • Prefer concrete actors, actions, comparisons, and consequences over abstract labels such as "the result," "the study," or "the advantage."
  • Keep the tone professional and restrained. Avoid hype, exaggerated urgency, empty drama, self-congratulation, and academic ceremony.
  • A strong paragraph usually contains a claim, the mechanism or evidence behind it, and the consequence for the reader.

Paragraphs And Sentences

  • Develop one coherent thought in a paragraph. Do not place unrelated fact cards next to each other.
  • Let paragraph and sentence length follow the thought. A page of equally sized paragraphs or uniformly short sentences feels mechanical.
  • Keep a condition, contrast, cause, and consequence connected when they form one idea. Split only at a genuine conceptual pause.
  • Do not repair choppy prose by joining unrelated facts with semicolons or colons.
  • Do not begin several adjacent paragraphs with the same abstract topic-sentence shape.
  • Do not replace one repeated pattern with another row of rhetorical questions, presenter cues, number-led openings, or forced transitions.
  • Avoid repeated openings such as "The study," "The result," "This shows," and "Furthermore" when a concrete subject can carry the sentence.
  • Do not build several adjacent claims from repeated negative constructions such as "is not," "does not," or 不是、没有、不意味着、并不、不能. State the positive requirement or relationship when that is what the reader needs.
  • Do not end a section by merely restating its first sentence.
  • Do not use em dashes in public prose.

Tables, Lists, And Diagrams

  • Use a table only when every row is compared on the same dimensions.
  • Do not use a table to compress explanations that readers need in prose.
  • Introduce what a diagram is meant to show before the diagram appears.
  • A list should make a real set or sequence visible. Do not turn every paragraph into bullets.
  • In a research article, one recurring scenario is often more useful than many disconnected examples.
Show full SKILL.md (615 more words)Show less

Recurring Failure Patterns

The kernel sees a write. The harness sees a tool call. Neither can decide.

This breaks one contrast-and-consequence chain into presentation notes. Keep the relationship connected when it is one thought.

研究给出了分类。结果很明显。这进一步说明了方案的优势。

These sentences have no concrete object or progression and could be reordered without changing their meaning.

并行 Agent 需要 contract-valid effect serializability。它还需要 semantic resource discovery、authority revalidation 和 global outcome contract。

This asks a Chinese reader to decode several undefined English abstractions before seeing the problem. Start with a concrete failure, explain what must be checked, and introduce a formal name only if it remains useful.

A sequence such as 跨事件策略反复出现为四类关系。、上下文依赖让强制执行更难 落地。、两类难点会叠加。 feels mechanical because adjacent paragraphs repeat one abstract opening. Any individual sentence may still be useful.

代码能编译,并不意味着它选对了目标,也不意味着它不会引入开销。

This makes the reader recover the real claim through repeated negation. When the point is a requirement, state it directly: 成功的调优还需要明确目标、控制开销, 并在测试中保持稳定。

Tone And Diction

  • Avoid empty judgments such as "a major breakthrough," "the advantage is clear," "further validates the superiority," and "truly changes everything."
  • Do not manufacture surprise, uncertainty, failure, or personal experience to make generated prose sound human.
  • State real limitations plainly, without apology or self-attack.
  • Prefer direct verbs and specific nouns. Cut filler such as "it is important to note that," "in order to," "due to the fact that," and "with respect to."
  • Do not use the tone of a peer-review response in a public article. The reader needs the argument, not ceremonial claims of novelty.

English

  • Use idiomatic technical English rather than paper-abstract phrasing.
  • Prefer active constructions when the actor matters, without forcing every sentence into active voice.
  • Avoid nominalized prose when a direct verb is clearer.
  • Keep modifiers close to the words they qualify.
  • Spell out the practical meaning of a formal property before naming it.
  • Use straight quotation marks for English prose and code literals.

Chinese

  • Write Chinese as Chinese. Do not preserve English sentence and paragraph boundaries or translate clause by clause.
  • Use Chinese for ordinary prose. English is appropriate for proper nouns, identifiers, code, paths, recognized terms of art without a clear translation, and useful search terms on first mention.
  • When a term has a clear Chinese explanation, give the Chinese explanation first and the English original in parentheses only when useful.
  • Do not write a Chinese sentence whose grammatical load is carried mostly by English nouns.
  • Use AI Agent or AI 智能体 consistently for the general role.
  • Put code, commands, paths, filenames, and identifiers in backticks.
  • Render table headers in Chinese except for proper nouns, identifiers, and established acronyms.
  • Render English quotations in natural Chinese unless exact wording matters.
  • Use full-width punctuation in Chinese prose and half-width punctuation inside code and identifiers.
  • Keep a half-width space between Chinese characters and Latin letters or digits, such as 64 个仓库 and eBPF 程序.
  • Avoid calques and repeated abstract subjects such as 论文、研究、这些发现 when the sentence can name the actual mechanism, workload, measurement, or consequence.
Chinese Style Anchor

Use this kind of Chinese technical rhythm as a positive anchor:

基于 Wasm,我们可以使用多种语言构建 eBPF 应用,并以统一、轻量级的方式管理和发布。以我们构建的示例应用 bootstrap.wasm 为例,大小仅为约 90K,很容易通过网络分发,并可以在不到 100ms 的时间内在另一台机器上动态部署、加载和运行,同时保留轻量级容器的隔离特性。运行时不需要内核头文件、LLVM、clang 等依赖,也不需要做任何消耗资源的重量级编译工作。

The example reads naturally because related clauses stay together, sentence length follows the thought, and each full stop lands after a complete claim. Treat it as a rhythm reference, not a required content pattern.

Final Reader Test

Before publishing, read only the title, description, opening, headings, and conclusion. A qualified reader should be able to answer:

  • What concrete problem does this article address?
  • Why do existing mechanisms leave a gap?
  • What decision or design does the article recommend?
  • Where does the recommendation stop applying?
  • Which one or two terms are worth remembering?

If those answers require reading the references or decoding several coined terms, rewrite the article.

© eunomia-bpf, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .agents/skills/blog-writing-style of eunomia-bpf/eunomia.dev.

Open the folder on GitHubat commit 4cc0b4f

Compare with similar skills

Blog Writing Style next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Blog Writing Style compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Blog Writing Style this skilleunomia-bpf/eunomia.dev236—~2.5kAutomated safety check: PassMIT
Readability Checkjdevalk/skills105—~3.3kAutomated safety check: PassMIT
SEO OptimizerOneWave-AI/claude-skills3361 repos~1.7kAutomated safety check: PassMIT
Content Productionborghei/Claude-Skills891—~4.8kAutomated safety check: PassMIT
SEO Content Writerthatrebeccarae/claude-marketing161—~1.1kAutomated safety check: PassMIT
Digen AILeoYeAI/openclaw-master-skills2.2k—~4.3kAutomated safety check: PassMIT

Similar skills

  • Readability Check

    jdevalk/skills

    Runs a readability audit on a blog post draft or other multi-paragraph prose, calibrated for readers who read English as a second language.

    105 GitHub stars~3.3k tokensUpdated 3 mo ago
    Writing & ContentAuto-check passed
  • SEO Optimizer

    OneWave-AI/claude-skills

    Optimize content for search engines with keyword analysis, readability scoring, meta descriptions, and competitor comparison.

    336 GitHub starsUsed in 1 repo~1.7k tokens
    Writing & ContentAuto-check passed
  • Content Production

    borghei/Claude-Skills

    Full content production pipeline from blank page to publish-ready piece: research, briefs, drafting, SEO, readability, and editorial gates.

    891 GitHub stars~4.8k tokensUpdated 4 days ago
    Writing & ContentAuto-check passed
  • SEO Content Writer

    thatrebeccarae/claude-marketing

    SEO-optimized content creation with brand voice analysis and platform-specific frameworks.

    161 GitHub stars~1.1k tokensUpdated 4 mo ago
    Writing & ContentAuto-check passed
  • Digen AI

    LeoYeAI/openclaw-master-skills

    digen-ai is a ClawHub skill that brings directed intelligent generation to your video projects.

    2.2k GitHub stars~4.3k tokensUpdated 2 mo ago
    Writing & ContentAuto-check passed
  • SEO Optimizer

    secondsky/claude-skills

    SEO optimization with keyword analysis, readability assessment, technical validation, content quality.

    227 GitHub stars~1.6k tokensUpdated 13 days ago
    Marketing & SEOAuto-check passed

More from eunomia-bpf/eunomia.dev

All 18 skills in this repo
  • Eunomia Community Radar

    eunomia-bpf/eunomia.dev

    Monitor approved eBPF, Linux, observability, and AI-infrastructure communities through watchlist-opted-in read-only Slack archives plus ordinary visible browser UI, turn the strongest daily…

    236 GitHub stars~2.5k tokensUpdated today
    Auto-check passed
  • Eunomia Research Report

    eunomia-bpf/eunomia.dev

    Research, write, validate, and publish source-grounded Eunomia Daily Reports for technical readers.

    236 GitHub stars~3k tokensUpdated today
    Auto-check passed
  • Juejin Publisher

    eunomia-bpf/eunomia.dev

    Prepare or publish eunomia.dev Markdown articles on Juejin. An agent skill from eunomia-bpf/eunomia.dev.

    236 GitHub stars~4.8k tokensUpdated today
    Auto-check passed
  • Content Launch Planner

    eunomia-bpf/eunomia.dev

    Plan contribution-first cross-platform launches for eunomia.dev content, GitHub projects, papers, tutorials, demos, and public updates.

    236 GitHub stars~1.9k tokensUpdated today
    Auto-check passed
  • Devto Publisher

    eunomia-bpf/eunomia.dev

    Prepare or publish eunomia.dev Markdown articles on DEV Community through faithful long-form syndication.

    236 GitHub stars~1.7k tokensUpdated today
    Auto-check: notes
  • Eunomia Social Radar

    eunomia-bpf/eunomia.dev

    Monitor the continuing public performance and conversation around Eunomia blogs, reports, projects, and platform posts.

    236 GitHub stars~1.6k tokensUpdated today
    Auto-check passed

Questions about Blog Writing Style

What does Blog Writing Style do?

Define the finished prose style for English and Chinese eunomia.dev blog posts and Research pages. dev.dev blog posts and Research pages.

When should I use Blog Writing Style?

Blog Writing Style fits situations like: reviewing reader comfort; research readability; recurring sentence-level AI tells.

How do I install Blog Writing Style in Claude Code?

Run `npx skills add eunomia-bpf/eunomia.dev --skill blog-writing-style -a claude-code`. Or copy the skill folder (.agents/skills/blog-writing-style in eunomia-bpf/eunomia.dev) into .claude/skills/blog-writing-style in your project. Claude Code loads it when a task matches its description.

How do I install Blog Writing Style in Codex?

Run `npx skills add eunomia-bpf/eunomia.dev --skill blog-writing-style -a codex`. Or copy the skill folder (.agents/skills/blog-writing-style in eunomia-bpf/eunomia.dev) into .agents/skills/blog-writing-style in your project. Codex loads it when a task matches its description.

Can I use Blog Writing Style in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add eunomia-bpf/eunomia.dev --skill blog-writing-style -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/blog-writing-style, .gemini/skills/blog-writing-style, .github/skills/blog-writing-style and .opencode/skills/blog-writing-style in your project.

What does Blog Writing Style need to run?

SKILL.md names no scripts, command-line tools or credentials: Blog Writing Style is instructions for the agent only.

Does Blog Writing Style access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Blog Writing Style safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Blog Writing Style use?

Blog Writing Style is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Blog Writing Style use?

About 2.5k tokens (SKILL.md is roughly 9.9k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Blog Writing Style?

Skills that share tags, products or a category with Blog Writing Style: Readability Check (jdevalk/skills, 105 stars), SEO Optimizer (OneWave-AI/claude-skills, 336 stars), Content Production (borghei/Claude-Skills, 891 stars) and SEO Content Writer (thatrebeccarae/claude-marketing, 161 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Blog Writing Style?

eunomia-bpf (a GitHub organization) maintains it in eunomia-bpf/eunomia.dev, which has 236 GitHub stars. The repository holds 18 skills in this directory. The repository was last updated on October 10, 2026.

Source: eunomia-bpf/eunomia.dev on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.