Agent skill

Swyx Writing

by swyxio in swyxio/skills

Apply swyx's reusable nonfiction writing preferences when drafting or materially revising prose, including technical posts, essays, research memos, reports, documentation, profiles, scripts…

MITAuto-check passedMedia & Creative

Install Swyx Writing

skills CLI
$ npx skills add swyxio/skills --skill swyx-writing -a claude-code

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

GitHub CLI
$ gh skill install swyxio/skills swyx-writing --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/swyxio/skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/swyx-writing .claude/skills/swyx-writing && 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
swyx-writing
GitHub stars
176
Token cost
~2.2k tokens
SKILL.md length
1,181 words
Files
3 (incl. references)
Skills in repo
89
Repo updated
First seen
Licence
MIT

At a glance

Apply swyx's reusable nonfiction writing preferences when drafting or materially revising prose, including technical posts, essays, research memos, reports, documentation, profiles, scripts…

  • Works in 3 steps: Developmental: check the reader change,… → Explanatory: check undefined nouns,… → Line: replace abstractions with actors…
  • Verbatim transcription
  • SKILL.md covers Start with the reader's change, Sound like swyx, Avoid generated prose and Attribute without writing…, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Swyx Writing is an agent skill from swyxio/skills. Apply swyx's reusable nonfiction writing preferences when drafting or materially revising prose, including technical posts, essays, research memos, reports, documentation, profiles, scripts, announcements, and publishing copy. Use as the shared style and editing layer alongside the skill that owns the subject or format. Do not use for verbatim transcription, code-only work, or fiction unless the user explicitly asks for swyx's nonfiction voice.

Its SKILL.md is about 2.2k tokens, which your agent loads only when the skill is triggered. The skill folder holds 4 other files, including reference files (for example `agents/openai.yaml` and `references/technical-writer-influences.md`).

It sits in Media & Creative, covering Creative writing and fiction and Transcription. The repository describes itself as: Agent skills for Claude Code and other AI agents. The licence is MIT.

When your agent uses it

  • Verbatim transcription
  • Fiction unless the user explicitly asks for swyxs nonfiction voice

Example prompts

  • “/swyx-writing”

Workflow steps

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

  1. Developmental: check the reader change, central question, order, strongest
  2. Explanatory: check undefined nouns, assumed knowledge, missing causal
  3. Line: replace abstractions with actors and verbs, vary rhythm naturally,

What it can do on your machine

Read from SKILL.md and the folder at commit 038ef34. 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

Swyx Writing loads about 2.2k tokens when it runs, and up to ~3.3k if it reads all its reference files. Until then it costs about 115 tokens; SKILL.md has 1,181 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~115
When it runs · the whole SKILL.md, loaded when a task matches
~2.2k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~3.3k

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 swyxio/skills at commit 038ef34, republished under its MIT licence (© swyxio). 1,181 words, ~2,180 tokens.

Download SKILL.mdSave it as .claude/skills/swyx-writing/SKILL.md (or your agent's skills folder). This skill also uses 2 other files; get the full folder from GitHub.
name
swyx-writing
description
Apply swyx's reusable nonfiction writing preferences when drafting or materially revising prose, including technical posts, essays, research memos, reports, documentation, profiles, scripts, announcements, and publishing copy. Use as the shared style and editing layer alongside the skill that owns the subject or format. Do not use for verbatim transcription, code-only work, or fiction unless the user explicitly asks for swyx's nonfiction voice.

swyx writing

Write so a smart reader can understand the point once, remember the useful model, and decide what to do next. Preserve the author's real judgment and the messiness that earned it. Do not imitate a generic founder, marketer, or technical-blog persona.

This skill owns voice, explanation, and editing. Pair it with the skill that owns the facts or format:

  • use research-grounded-writing when claims require external or primary evidence;
  • use ai-devblog for technical story selection and publication;
  • use ai-readme for executable repository documentation;
  • use person-profile-writing for biographies and identity research;
  • use video-talk-to-essay for recorded-talk articles.

The genre skill decides what belongs. This skill decides how the prose reads.

Start with the reader's change

Before substantial drafting, identify privately:

  • what the reader currently believes, misunderstands, or cannot do;
  • what they should believe, understand, feel, or try afterward;
  • the one question the piece answers;
  • the strongest fact, example, or tension that earns their attention;
  • tempting material that belongs somewhere else.

Name the actual subject and stakes early. Do not hide the lede behind a mystery hook, abstract principle, slogan, or narration about the research process. A clever line may sharpen a clear idea; it may not replace one.

Prefer one developing example over a sequence of disconnected examples. Give the reader the observable consequence, then the smallest concrete example, then a plain model. Add formal terms, implementation detail, and edge cases only when they help the intended reader.

Sound like swyx

Use plain words around precise ideas. Prefer concrete subjects and active verbs: name the person or system doing the work. Technical terms should earn their place by naming something accurately, not by making the surrounding prose sound technical.

Let voice come from real judgment:

  • state the useful opinion and the reason for it;
  • keep authentic surprise, disagreement, cost, mistakes, and reversals;
  • explain tradeoffs without sanding them into neutral mush;
  • use humor, analogy, fragments, and asides when they clarify or control pace;
  • widen a claim only as far as the evidence or experience earns;
  • distinguish shipped behavior, proposals, observations, and inference.

Give the reader one manageable thought at a time. Let longer sentences establish circumstances or mechanisms. Let short sentences land a discovery or judgment. Use paragraph breaks where a person would naturally pause. Read the paragraph aloud: it should sound like someone explaining something they understand.

Adapt tone and storytelling to the content and the reader's purpose. A career profile may develop through transitions and defining contributions; a technical explanation may follow a mechanism and its tradeoffs; a practical guide may organize around decisions and steps. Choose the approach that makes this subject clear and engaging. Do not force every piece into the same narrative arc or voice. Let the evidence earn the drama, humor, and degree of informality.

Avoid generated prose

Remove patterns that make the writing feel produced by a template:

  • repeated not X, but Y, X is not Y, and perfectly balanced reversals;
  • uniform triads, symmetrical sections, bold thesis restatements, and mandatory recaps;
  • every heading, paragraph, and sentence competing to be an aphorism;
  • every observation enlarged into a universal principle;
  • fake quotations, invented reactions, and a suspiciously clean causal history;
  • generic transitions such as the broader principle, the common thread, it is worth noting, in today's landscape, or this underscores;
  • delve, leverage, foster, robust, seamless, and other prestige words when an ordinary verb says more;
  • authority, boundary, contract, receipt, durable, surface, and exact used as atmosphere rather than necessary terms;
  • noun piles and internal project vocabulary before the concrete behavior;
  • exhaustive evidence, implementation inventories, or caveats in the main narrative merely because they are available;
  • a second conclusion that repeats the first in summary language.

Do not solve these problems by making the prose flat. Earn one or two memorable sentences by compressing a true, useful distinction. Keep idiosyncratic phrasing when it sounds intentional and remains easy to understand.

Show full SKILL.md (539 more words)Show less

Attribute without writing about attribution

Put links and evidence beside the claim they support. Make the person, system, event, or idea the grammatical subject whenever possible. Prefer The release controller deploys the Worker to The evidence establishes a release authority boundary.

Do not begin sentences with According to, The documentation says, or The available evidence shows unless the source itself is the subject. State the supported fact directly and link the useful words. Preserve uncertainty where the source does not justify a clean assertion.

For example, prefer X founded Y to build inference software to The conference biography identifies X as the founder of Y, when the role is supported. A company profile explains its products and work; the supplied recording does not establish broader archive coverage belongs in an internal coverage note. A biography selects what a technical contribution reveals about the person; the presentation’s sequence of tuning steps belongs on the talk page. These are selection examples, not fixed sentence templates or permission to remove material uncertainty.

Format for the argument

Use connected prose to develop an explanation, but do not let long stretches run without headings. Give substantial pieces descriptive section headings at meaningful turns so readers can scan, navigate, and return to a point. As a guideline, reconsider the structure after several substantial paragraphs or roughly 300–500 words without a heading; this is a readability cue, not a quota. Short pieces may not need headings. Name what the section explains rather than using generic labels or forcing a fixed outline.

Actively look for opportunities to present distinct contributions, ideas, options, examples, or lessons as bullets under a useful section heading. Begin each bullet with a descriptive bold label when it helps readers find the point, then explain it with enough context, mechanism, or example to be useful. Keep chronology and causal development in prose when splitting them would weaken the story. Use numbered lists for real sequences, tables for repeated-field comparisons, and visuals when they reveal a relationship that prose makes hard to inspect. Choose structure for this content; avoid uniform bullet counts, symmetrical sections, and inventories that replace explanation.

Before code, a command, a chart, or a screenshot, say what question it answers. Afterward, interpret what matters. Artifacts should advance the reader's model, not serve as proof that work happened.

Edit in three passes

  1. Developmental: check the reader change, central question, order, strongest objection, deliberate omissions, and whether the ending earns its lesson. Check that tone and storytelling suit the subject, long prose has useful headings, and labeled bullets improve genuinely parallel material.
  2. Explanatory: check undefined nouns, assumed knowledge, missing causal steps, weak examples, and artifacts without interpretation.
  3. Line: replace abstractions with actors and verbs, vary rhythm naturally, remove repeated antithesis and conclusions, and cut anything that supplies neither information, judgment, voice, nor pace.

Factual verification is separate from the editorial passes. When confusion risk is high, give an uninvolved reader only the draft and ask what it is about, what changed, how the central mechanism works, what supports it, and what remains uncertain. Fix the prose rather than coaching the reader.

For substantial public technical writing, read technical-writer influences. Use the writers' decisions as inspiration; never imitate their surface persona.

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

Files

SKILL.md and 2 other files (references) in swyx-writing of swyxio/skills.

  • SKILL.md
  • agents/openai.yaml
  • references/technical-writer-influences.md

Open the folder on GitHubat commit 038ef34

Compare with similar skills

Swyx Writing 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.

Swyx Writing compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Swyx Writing this skillswyxio/skills176—~2.2kAutomated safety check: PassMIT
HyperFrames Media Useheygen-com/hyperframes60k—~2.4kAutomated safety check: PassApache-2.0
Native Subtitle Quote Imagechengyi-ai/native-subtitle-quote-image2.6k—~2.4kAutomated safety check: PassMIT
Edu Chem Videowy51ai/edulab1.4k—~2.1kAutomated safety check: NotesApache-2.0
Transcription Memory ReconstructionNxcoreAI/EverRoom3k—~714Automated safety check: PassCustom licence
Edu Math Videowy51ai/edulab1.4k—~2.5kAutomated safety check: NotesApache-2.0

Similar skills

  • HyperFrames Media Use

    heygen-com/hyperframes

    Finds, generates and edits media for HyperFrames video projects: music, sound effects, images, icons, logos, voiceovers, captions and color grades.

    60k GitHub stars~2.4k tokensUpdated today
    Media & CreativeAuto-check passed
  • Native Subtitle Quote Image

    chengyi-ai/native-subtitle-quote-image

    将本地视频或用户有权处理的在线视频,经过来源获取、文字稿定位、选题选句、精确取帧、紧凑裁切、拼图和逐张质检,制作成 3:4 或保留画面原比例的视频字幕长图。支持两种明确分开的输出:保留画面内已烧录字幕的原生字幕模式,以及把已审核的时间点与台词绘制到真实视频帧上的脚本字幕模式。用户要求原生字幕截图、字幕帧拼图、YouTube…

    2.6k GitHub stars~2.4k tokensUpdated today
    Media & CreativeAuto-check passed
  • Edu Chem Video

    wy51ai/edulab

    A skill your agent uses when asked to make an explainer / walkthrough video (讲解视频、解题视频、例题精讲、微课) for a chemistry problem (化学题: 氧化还原配平 双线桥 电子守恒, 物质的量计算, 化学平衡 三段式 平衡常数 转化率 反应速率, 离子反应, 电化学, 溶液 滴定…

    1.4k GitHub stars~2.1k tokensUpdated yesterday
    Media & CreativeAuto-check: notes
  • Reconstruct a complete, searchable memory from an untrusted meeting or conversation transcript.

    3k GitHub stars~714 tokensUpdated 2 days ago
    Media & CreativeAuto-check passed
  • Edu Math Video

    wy51ai/edulab

    A skill your agent uses when asked to make an explainer / walkthrough video (讲解视频、解题视频、例题精讲、微课) for a math problem (数学题, geometry, algebra, functions, motion/行程 problems), from a problem screenshot…

    1.4k GitHub stars~2.5k tokensUpdated yesterday
    Media & CreativeAuto-check: notes
  • Transcribe

    JetBrains/skills

    Official

    Transcribe audio files to text with optional diarization and known-speaker hints.

    366 GitHub starsUsed in 4 repos~776 tokens
    Media & CreativeAuto-check passed

More from swyxio/skills

All 89 skills in this repo
  • Programmatic Agents

    swyxio/skills

    Run a selected coding-agent CLI programmatically, with latency, error, usage, cost, and trace logging.

    176 GitHub stars~2.2k tokensUpdated today
    Auto-check passed
  • Design, implement, audit, or refresh protected username and handle namespaces for public products.

    176 GitHub stars~1.1k tokensUpdated today
    Auto-check passed
  • New Mac Setup

    swyxio/skills

    Fully automated new Mac setup for fullstack web developers and AI engineers.

    176 GitHub stars~4.3k tokensUpdated today
    Auto-check passed
  • Youtube API

    swyxio/skills

    Manage YouTube videos programmatically via the YouTube Data API v3 — upload video files, upload custom thumbnails, update video metadata (titles, descriptions, tags), and query video/channel info…

    176 GitHub stars~2.2k tokensUpdated today
    Auto-check passed
  • Batch YouTube Studio upload workflow for videos sourced from Airtable, Google Drive, Loom, YouTube, or local files.

    176 GitHub stars~1.5k tokensUpdated today
    Auto-check: warnings
  • Reconstruct and visually analyze paired agent, game, or policy trajectories to determine whether changed actions produced their intended effects.

    176 GitHub stars~1.8k tokensUpdated today
    Auto-check passed

Questions about Swyx Writing

What does Swyx Writing do?

Apply swyx's reusable nonfiction writing preferences when drafting or materially revising prose, including technical posts, essays, research memos, reports, documentation, profiles, scripts…. Swyx Writing is an agent skill from swyxio/skills. Apply swyx's reusable nonfiction writing preferences when drafting or materially revising prose, including technical posts, essays, research memos, reports, documentation, profiles, scripts, announcements, and publishing copy.

When should I use Swyx Writing?

Swyx Writing fits situations like: verbatim transcription; fiction unless the user explicitly asks for swyxs nonfiction voice.

How do I install Swyx Writing in Claude Code?

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

How do I install Swyx Writing in Codex?

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

Can I use Swyx Writing 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 swyxio/skills --skill swyx-writing -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/swyx-writing, .gemini/skills/swyx-writing, .github/skills/swyx-writing and .opencode/skills/swyx-writing in your project.

What does Swyx Writing need to run?

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

Does Swyx Writing 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 Swyx Writing 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 Swyx Writing use?

Swyx Writing 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 Swyx Writing use?

About 2.2k tokens (SKILL.md is roughly 8.7k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 1.1k tokens, read only when the agent opens those files.

What are the alternatives to Swyx Writing?

Skills that share tags, products or a category with Swyx Writing: HyperFrames Media Use (heygen-com/hyperframes, 60k stars), Native Subtitle Quote Image (chengyi-ai/native-subtitle-quote-image, 2.6k stars), Edu Chem Video (wy51ai/edulab, 1.4k stars) and Transcription Memory Reconstruction (NxcoreAI/EverRoom, 3k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Swyx Writing?

swyxio (a GitHub user) maintains it in swyxio/skills, which has 176 GitHub stars. The repository holds 89 skills in this directory. The repository was last updated on October 5, 2026.

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