agentleFS
Sign inStart free
javierbreajavierbrea/eslint-plugin-boundaries/.agents/skills/writing-style/SKILL.md

writing-style

Skill1k starsChanged 11 days agoMITView on GitHub

Style for everything written for people on GitHub or in plans — issue bodies, pull request descriptions, RFCs, review and issue comments, release notes and the files under .agents/plans/. Load before drafting any of them, and before reviewing one for style.

We split this skill into the rules it gives an agent

7 rules. 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 27 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use a bulleted list for parallel items (criteria, steps, files, options, findings) and prose for an argument.

    line 14
  2. Distinctive:

    Use a numbered list only when the order matters.

    line 15
  3. Distinctive:

    Use a table to compare things along two or more attributes.

    line 16
  4. Very specific:

    Use a checklist (- [ ]) only for items that get ticked off: acceptance criteria, task lists.

    line 17
  5. Distinctive:

    Do not hard-wrap paragraphs.

    line 19
  6. Very specific:

    No emphasis for urgency (IMPORTANT, bold sentences, exclamation marks) and no emojis.

    line 27
  7. Distinctive:

    Delete a sentence that does not change what the reader does or decides.

    line 35

“Common” is measured: kept by at least 6 in 10 of the 27 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

Writing style

Everything published from this repository is read by someone who was not in the session: a contributor picking up an issue, a reviewer, the community reading an RFC, or an agent implementing a plan with no other context. Write for that reader.

Shape

  • Lead with what the reader needs to know or decide. Context comes second, history last or not at all.
  • One idea per paragraph, and a paragraph is at most four or five lines. Anything longer is split or becomes a list.
  • Use a bulleted list for parallel items (criteria, steps, files, options, findings) and prose for an argument. A bullet is one or two sentences; a bullet that needs a paragraph is a section.
  • Use a numbered list only when the order matters.
  • Use a table to compare things along two or more attributes. Three rows or fewer with one attribute is a list, not a table.
  • Use a checklist (- [ ]) only for items that get ticked off: acceptance criteria, task lists. Never for prose.
  • Headings (##, ###) name what the section holds (Context, Proposal, Acceptance criteria), not what the writer did. A comment or another text under about fifteen lines has no headings, unless its template or skill sets them (issues, pull requests).
  • Do not hard-wrap paragraphs. One paragraph is one line; blank lines separate paragraphs. GitHub renders single line breaks as breaks.

Register

  • English, whatever language the conversation is in, plain and direct. State facts and decisions: "the playground is its own package", not "we could consider making the playground its own package". A real open question is written as a question under Open questions.
  • Say what and why. Leave out how the text was produced ("I checked", "after looking at", "it seems that").
  • Be concrete: name the file, the command, the number, the issue. #N for issues and pull requests, the full URL for a discussion, backticks for paths, commands, options and code, a fenced block for anything multi-line.
  • Link instead of restating. One link per fact, placed where the fact is used. When a fact already lives in a repository file, point to the file.
  • No emphasis for urgency (IMPORTANT, bold sentences, exclamation marks) and no emojis. The only exception is the status marker on the first line of an RFC (**Status:** 🟡 Draft, **Status:** ✅ Accepted (2026-10-25)).
  • Expand an acronym the first time it appears unless it is a common one in this repository (ESLint, RFC, PR, CI, URL).

Length

  • An issue says what, why and how it is accepted. It fits on one screen when the task is one day's work. Details that only the implementer needs go to the plan or to the pull request.
  • A pull request description says what changed for its users and what the reviewer must look at. Evidence (benchmarks, screenshots, profiles) goes in its own section after the summary.
  • A plan file is as long as the task needs, but it points at repository files instead of copying them.
  • Delete a sentence that does not change what the reader does or decides.

Per artifact

ArtifactReaderStructure
IssueA contributor or agent picking it upcreate-issue skill
Pull requestA reviewer.github/PULL_REQUEST_TEMPLATE.md, filled by the create-pr skill
RFCThe community, before the decisiondocs/contributing-workflow.md, step 1
PlanAn agent with no other contextwrite-plan skill
Changelog entryA user of the package.agents/rules/changelog-and-versioning.md, "Entry style"
Review or issue commentThe author of the changeFinding first, then the reason, then the suggestion; one comment per finding

Download now (ZIP)

  • Other versions28 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 28

    1. pydantic/monty★ 8.8ksame name
    2. astashov/liftosaur★ 768same name
    3. open-thoughts/OpenThoughts-Agent★ 301same name
    4. anotb/management-consulting-plugin★ 62same name
    5. kaustubhhiware/agent-skills-fullstack-development★ 16same name
    6. lout33/writing-style-skill★ 10same name
    7. chnlich/charlie-bot★ 9same name
    8. Isaac-Flath/agent-starter-skills★ 8same name
    9. coddingtonbear/skills★ 3same name
    10. PatriotAgenticLLC/better_claude_skillz★ 3same name

    18 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.

  • Used alongsideProjects with it also use changeset-scope, skill-prefixes and 4 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.
    • timmo001Skill

      changeset-scope

      Keep all scoped code work contained to the user-defined changeset. Use for implementation, fixes, diagnosis, refactoring, cleanup, and review when explicit instructions, named files, diffs, branches, pull requests, or injected work scopes define the boundary.

      Also in 3 projects that use this

    • timmo001Skill

      skill-prefixes

      Name skills with an owner prefix so they do not collide with shared, imported or other repositories' skills. Use when creating, renaming, importing or reviewing a skill name, especially for repository-local or tool-owned skills.

      Also in 3 projects that use this

    • NVIDIASkill

      testing

      Run Python and Rust tests for physicsnemo-curator using uv + pytest with coverage reporting via pytest-cov, and cargo-nextest for Rust tests. Includes benchmark workflows with pytest-benchmark and criterion.

      Also in 3 projects that use this

    • GoogleSkill

      code-review

      Adversarial systems, concurrency, memory-model, and architecture code review protocol. Use this skill whenever asked to review code, PRs, diffs, or branches in Filament.

      Also in 3 projects that use this

    • +2 more from timmo001
  • In this repository12 other files it gives its agents.

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.

Check your setup, free