agentleFS
zhamannzhamann/skills/write/SKILL.md

write

Skill2 starsChanged 28 days agoMITView on GitHub

Write prose people want to read, with a real point of view and a voice suited to its audience. Use for articles, essays, guides, explanations, emails, posts, documentation, stories, and other writing, whether drafting from scratch or revising existing text. Not for code or text whose wording must stay exact.

35 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use read, hunt and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • MicrosoftSkill

    read

    Read pages from a document. Returns MinerU markdown (preferred) or PyPDF text, with optional page screenshots and sub-image fetching. When markdown is available, a figure_images_meta catalog is included — use the figures parameter to fetch specific sub-image pixels. Triggers include "read page 5", "what does page 12 say", "show me the chart on page 3", "extract pages 3-5".

    Also in 6 projects that use this

  • ninehillsSkill

    hunt

    Finds root cause before applying fixes for errors, crashes, regressions, failing tests, broken behavior, and screenshot-reported defects. Use when users report in any language errors, crashes, broken behavior, regressions, failing tests, screenshot evidence, or something that used to work and now fails. Not for code review or new features.

    Also in 4 projects that use this

  • vltingSkill

    think

    Lightweight plan-to-execution pipeline. Explores codebase, runs council deliberation, presents plan for approval, then executes via /run.

    Also in 7 projects that use this

  • genomewalkerSkill

    health

    Soul system health check with remediation

    Also in 5 projects that use this

  • AnthropicSkill

    analyze

    Answer data questions -- from quick lookups to full analyses. Use when looking up a single metric, investigating what's driving a trend or drop, comparing segments over time, or preparing a formal data report for stakeholders.

    Also in 6 projects that use this

  • anombyte93Skill

    discover

    Phase 1 of the prd-taskmaster pipeline: brainstorm-driven discovery. Delegates to superpowers:brainstorming in Interactive Mode (one adaptive question at a time), or self-brainstorms in Autonomous Mode when no user is present. Intercepts before the brainstorming chain hands off to writing-plans — this skill owns the exit. Extracts constraints, calibrates scale (Solo / Team / Enterprise), and advances the pipeline to GENERATE.

    Also in 4 projects that use this

  • Other versions35 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 35

    1. reactjs/react.dev★ 12ksame name
    2. tw93/claude-health★ 7.5ksame name
    3. tw93/Waza★ 7.5ksame name
    4. docker/docs★ 4.7ksame name
    5. waynesutton/markdown-site★ 628same name
    6. ohad6k/emulo★ 292same name
    7. ninehills/skills★ 280same name
    8. hsantanna88/clo-author★ 262same name
    9. hugosantanna/clo-author★ 262same name
    10. yan-labs/yan-skills★ 232same name

    25 more not listed

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository3 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 31 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write something people want to read: a piece with a person behind it, someone who knows the subject, has a view on it, and is talking to a …

    line 10
  2. Distinctive:

    Specificity without fabrication: Use concrete examples, numbers, comparisons, and named sources when they're known and useful.

    line 22
  3. Distinctive:

    Plain, not flat: Prefer familiar words and direct constructions, and cut filler, inflated phrasing, promotional adjectives, and empty …

    line 25
  4. Distinctive:

    Judgment over formulas: No word, punctuation mark, sentence length, grammatical voice, or format is wrong in itself.

    line 27
  5. Distinctive:

    End when done: Stop once the piece has done its job.

    line 35
  6. Distinctive:

    Leave what already works. Don't replace their style with yours.

    line 45

“Common” is measured: kept by at least 6 in 10 of the 31 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Write

Objective

Write something people want to read: a piece with a person behind it, someone who knows the subject, has a view on it, and is talking to a particular reader.

Accuracy and clarity are the floor, not the goal. Writing fails in two directions. One is performance: hype, filler, empty enthusiasm. The other is inert neutrality: correct, careful, and lifeless, like a reference entry nobody chose to read. Avoiding the first must not produce the second.

Good writing doesn't come from avoiding a list of words or imitating human quirks. It comes from having something to say, knowing who it's for, and caring whether it lands.

Core principles

  • Purpose before prose: Know what the piece must communicate, to whom, and what the reader should understand, feel, or do afterward.
  • A writer, not a reference: Interpret instead of only reporting. Say what matters, what's misleading, what you'd do, and why. Readers come for judgment they can't get from a spec sheet. A view presented as the writer's own is not a fabrication.
  • The reader's situation: Write to someone in a particular moment: what they're looking at, what's bothering them, what they'd ask next. Ground explanations in that moment with a scenario, a decision, or a mistake they're about to make. A clearly hypothetical example is illustration, not invention.
  • Substance over performance: Build on claims, observations, mechanisms, examples, and consequences. Personality is not performance; filler and hype are.
  • Specificity without fabrication: Use concrete examples, numbers, comparisons, and named sources when they're known and useful. Never invent facts, experiences, quotations, or sources. When support is missing, qualify the claim once, find support if the task allows, or cut it.
  • Commitment: State real uncertainty once, where it applies, then speak plainly. Reflexive qualifiers drain authority without adding accuracy.
  • Voice from context: Choose vocabulary, rhythm, formality, and perspective for the subject, audience, and medium. A support reply, a technical guide, and a personal essay need different voices, but none should sound like nobody wrote it. When writing as or for someone, match their voice.
  • Plain, not flat: Prefer familiar words and direct constructions, and cut filler, inflated phrasing, promotional adjectives, and empty transitions. Keep the color that makes a point land: a vivid comparison, a blunt verdict, a dry aside.
  • Structure from meaning: Organize around the ideas and the reader's needs, not a reusable template. Use the natural number of points. Give important ideas room and minor ones less.
  • Judgment over formulas: No word, punctuation mark, sentence length, grammatical voice, or format is wrong in itself. Use whatever serves the piece.

Workflow

  1. Establish direction. Determine purpose, audience, medium, tone, and constraints. Picture the reader: why they're here, what they already know, what they hope to get. For straightforward tasks, infer reasonable defaults instead of interrupting to ask.
  2. Decide what you think. Find the central point and your take on it: what a knowledgeable friend would tell the reader that a manual wouldn't. Then gather the support and distinctions it needs.
  3. Choose the form. Open where the reader's interest is: a claim, a question, a scene, a problem they recognize. Skip ceremonial introductions and definitions nobody asked for.
  4. Compose. Develop ideas through explanation, evidence, example, or narrative as the subject demands. Each paragraph should advance the piece, not restate the one before it. Vary sentence and paragraph length with emphasis and complexity. Use headings, lists, and emphasis only when they help readers navigate.
  5. End when done. Stop once the piece has done its job. Skip the summary, uplifting takeaway, or invitation to continue unless it adds something.
  6. Review the draft as the reader. Cut what's generic, inflated, redundant, or unsupported. Find where it goes flat: passages that report without saying why it matters, hedge without cause, or could have come from a manual. Give them a point of view or cut them. Watch for repeated shapes that make prose sound generated: the same contrast construction, reflexive three-item lists, stock transitions, matching sentence openings, a tidy closing line.

Revising someone else's writing

The same principles apply, with one more constraint: the piece belongs to its author.

  • Identify what they're trying to say and what must stay unchanged.
  • Preserve their meaning, factual claims, level of certainty, perspective, and recognizable voice.
  • Rewrite at the level of meaning: clarify what each passage is trying to say, then say it naturally. Treat surface patterns as clues, not automatic violations.
  • Leave what already works. Don't replace their style with yours.

Checks

  • Is there a person behind this, or could it pass unchanged as a manual or encyclopedia entry?
  • Would the intended reader keep going past the first paragraph? Why?
  • Does it say what the writer thinks, not only what is true?
  • Could this paragraph appear unchanged in an unrelated piece? If so, make it specific or cut it.
  • Are facts and uncertainty represented accurately, with nothing invented?
  • Would removing anything make the piece better?

Guiding maxim: Have something to say, say it like you mean it, and stop when it's said.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.