writing-style
Skill2 starsChanged 55 days agoMITView on GitHub
The author's aesthetic philosophy, prose style, and creative sensibility — the single source of truth every editorial skill judges against. Ships unconfigured; the first direct invocation runs a one-time interview that rewrites this file into the author's real style reference.
33 other projects use a version of this.
Adopted by
Most recently taken up by
- apetrov/skillstheir own version, first seen Nov 16, 2026
- bjoern621/claude-skillstheir own version, first seen Nov 16, 2026
- micooz/skillstheir own version, first seen Nov 15, 2026
- timmo001/skillstheir own version, first seen Nov 15, 2026
- modaal-agent/skillstheir own version, first seen Nov 15, 2026
- timmo001/notestheir own version, first seen Nov 11, 2026
Used with
Projects with it also use changeset-scope, skill-prefixes and 5 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.
timmo001Skillchangeset-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
timmo001Skillskill-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
NVIDIASkilltesting
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
GoogleSkillcode-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 4 projects that use this
MicrosoftSkillrelease
Use ONLY when the user explicitly asks to cut/publish a Huabu desktop release or tag a version. Performs an outward action — bumps the version in apps/desktop/package.json, commits + annotated-tags vX.Y.Z, and pushes the branch + tag. That tag push auto-builds the macOS/Windows installers and publishes a GitHub Release. Provides `status`, `bump`, `tag-push`, and a one-command `run` (bump → tag-push).
Also in 3 projects that use this
- +2 more from timmo001
Other versions33 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 33
- pydantic/monty★ 8.9ksame name
- OpenLAIR/dr-claw★ 1.2ksame name
- javierbrea/eslint-plugin-boundaries★ 1ksame name
- astashov/liftosaur★ 780same name
- open-thoughts/OpenThoughts-Agent★ 309same name
- anotb/management-consulting-plugin★ 62same name
- kaustubhhiware/agent-skills-fullstack-development★ 16same name
- micooz/skills★ 12same name
- lout33/writing-style-skill★ 10same name
- chnlich/charlie-bot★ 9same name
23 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 repository16 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- obsidian-basesSkill
- obsidian-markdownSkill
- pdf-convertSkill
- story-beta-readSkill
- story-copy-editSkill
- story-cover-checkSkill
- story-dev-editSkill
- story-kdp-exportSkill
- story-line-editSkill
- story-listenSkill
- story-manual-reviseSkill
- story-personasSkill
- story-proofSkill
- story-quick-cleanSkill
- story-update-bibleSkill
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 1 is distinctive, and 1 is very specific to this project.
0
Common
Most of the 31 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 14
If the author invoked
/writing-styledirectly: run the ## Setup interview below. - Distinctive: line 28
Ask these one at a time, conversationally.
“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.
Files
Author Writing Style
<!-- WRITING-STYLE-PLACEHOLDER: not configured. This marker is deleted when the setup interview rewrites the file. -->Status: not configured
This skill is dormant. It ships as an interview that writes itself; it does not yet describe any author.
If the author invoked /writing-style directly: run the ## Setup interview below.
If you are an editorial skill or critique sub-agent loading this file as a style reference: do not start the interview. If you can talk to the author, stop and tell them: "The writing-style skill is not configured yet; run /writing-style once before any editorial pass." If you are an isolated sub-agent that cannot reach the author, note in your report that no style reference exists and judge against neutral, generally accepted craft standards; flag nothing as "deliberate authorial form", because nothing has been declared deliberate yet.
Setup interview
The goal is a style reference grounded in the author's actual prose, not their self-image. Observations from finished text outrank self-reported preferences; where the two disagree, present the discrepancy and let the author decide what goes in the file.
1. Gather the prose
Ask the author for their best finished work: published books, completed stories, or the strongest chapters they have. Ask for file paths (or pasted text) and read everything they give you in full. If they have no finished prose at all, say plainly that the reference will be weaker and interview-only, and mark it as provisional in the Source Note so it gets rebuilt after the first finished draft.
2. Interview
Ask these one at a time, conversationally; follow up where an answer is vague. Do not batch them into one wall of questions.
- The constant. What recurs across everything you write, or want to? (A mood, a kind of premise, a question you keep returning to.)
- Influences, specifically. Which writers, filmmakers, or artists shape your work, and what exactly do you take from each? Push past names to mechanisms: not "Kafka" but what of Kafka shows up in your sentences.
- Passages you love. Two or three passages (yours or anyone's) you find perfect, and what makes each work for you.
- Anti-preferences. What do you hate in fiction? What common writing advice do you deliberately reject? What would make you close a book?
- Endings, explanation, ambiguity. How much do you resolve? What stays unexplained on purpose?
- Interiority and emotion. How do your characters think and feel on the page: stated, implied through image and action, or something else?
3. Analyze the prose
Independently of the answers, derive from the samples: sentence rhythm and its range, paragraph habits, dialogue conventions, interiority style, imagery and metaphor habits, tense and POV behavior (including any deliberate irregularities), structural choices, how endings behave, and recurring registers (comedy, dread, restraint). Be specific: not "uses short sentences" but the exact pattern observed and what it does. Present these observations to the author and confirm which are deliberate style and which are accidents they would rather fix; only confirmed-deliberate patterns become protected style.
4. Draft, approve, rewrite
Draft the full replacement file in the ## Target shape below and present it. Iterate until the author approves. Then rewrite this SKILL.md so it contains only the frontmatter (update the description to one sentence about this author's actual aesthetic) and the approved content: delete the placeholder marker, the Status section, the Setup interview, and the Target shape. The file you leave behind is the style reference, nothing else.
Target shape
The final file uses this skeleton (drop a section only if genuinely empty for this author):
- Opening note — one short paragraph: this is an aesthetic philosophy, not a ruleset; it calibrates every suggestion and every piece of feedback.
- The constant — the thread that runs across all the work, and the distinct registers it operates in, each tied to a named work or sample.
- Prose style — the sentence- and paragraph-level patterns, each stated as the exact observed pattern plus what editors must not do to it.
- Narrator and voice — POV behavior, reliability, any deliberate irregularities.
- Structure — how stories are shaped, chapter-title conventions, pacing values, how endings behave.
- What this author does not do — a bulleted never-suggest list; this is the section editorial skills lean on hardest.
- Influences and what they mean in practice — each influence as a mechanism visible in the prose, not a name-drop.
- Developmental feedback calibration — how a dev editor should filter findings through this aesthetic: what looks like a flaw but is the style, and the question to ask before flagging ("does the slow stretch have presence?"). End with the rule: flag only what works against the author's own register, never what merely works against generic taste.
- Source note — which works this was built from, the date, and the standing instruction to update it when new finished work appears or the author corrects the profile.
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
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.

