writing-style
Skill0 starsChanged 30 days agoApache-2.0View on GitHub
Revise prose so it reads as though a person wrote it, without the tells that give an AI draft away. Use it whenever prose is about to leave your hands, not only when someone asks for it: PR descriptions and PR review comments, issue bodies, commit message bodies, README and documentation sections, design-doc prose, and the summary or explanation you are about to write back to the user. Also use it when someone asks to improve, humanize, clean up, tighten, or revise a draft, or says something sounds robotic or AI-written. Apply it proactively — the drafts that need it most are the ones nobody thought to ask about, and a repo can declare its own terms of art that generic advice would strip out. It is about how the prose reads, not what it says: it does not decide a document's content or supply what the document is missing.
29 other projects use a version of this.
Adopted by
Most recently taken up by
- timmo001/notestheir own version, first seen Nov 8, 2026
- timmo001/contexttheir own version, first seen Nov 8, 2026
- OpenLAIR/dr-clawtheir own version, first seen Nov 8, 2026
- timmo001/skillstheir own version, first seen Nov 7, 2026
- javierbrea/eslint-plugin-boundariestheir own version, first seen Nov 2, 2026
- TheMarnix/agent-config-kittheir own version, first seen Nov 1, 2026
Used with
Projects 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.
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
- 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
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
- +2 more from timmo001
Other versions29 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 29
- pydantic/monty★ 8.9ksame name
- OpenLAIR/dr-claw★ 1.2ksame name
- javierbrea/eslint-plugin-boundaries★ 1ksame name
- astashov/liftosaur★ 775same name
- open-thoughts/OpenThoughts-Agent★ 309same name
- anotb/management-consulting-plugin★ 62same name
- kaustubhhiware/agent-skills-fullstack-development★ 16same name
- lout33/writing-style-skill★ 10same name
- chnlich/charlie-bot★ 9same name
- Isaac-Flath/agent-starter-skills★ 8same name
19 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 repository23 other files it gives its agents.
The file
We split this skill into the rules it gives an agent
2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.
0
Common
Most of the 28 edited versions we read keep these.
2
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
- Distinctive: line 21
When invoked directly with a draft: identify patterns below, revise, return the revised text.
- Distinctive: line 47
Forced rule of three: Use the actual count.
“Common” is measured: kept by at least 6 in 10 of the 28 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
!${CLAUDE_PLUGIN_ROOT}/scripts/skill-preflight.sh writing-style 2>&1 || true
When invoked directly with a draft: identify patterns below, revise, return the revised text. When producing prose: apply these patterns from the start rather than producing and then revising.
The rules live in rules.yaml
skills/writing-style/rules.yaml is the single authoritative source. Read it
for the banned words, phrases, and frequency thresholds. It carries the terms
and the reason for each, grouped by category.
A repository can declare terms of art the rules must not fire on, through a
## Prose Vocabulary: header in its CLAUDE.md. Honor a repository's
declaration when you draft in it: shirabe declares tier, journey, and
underscore because those are its own vocabulary, not tells.
What no matcher reaches
The judgment_only section of the rule source holds what no matcher
decides, and it is where the value is:
- Low information density. Well-formed sentences that say nothing. A fluent, entirely vacuous document produced ten alerts under three off-the-shelf style packages and not one concerned the vacuity.
- Empty conclusions. A closing paragraph that adds no content.
- Demonstratives with no antecedent. "This" and "that" pointing at nothing nameable.
- Attribution without a citation. "Studies show" with no study.
- Synonym cycling. Repeat the word instead.
- Forced rule of three. Use the actual count.
from X to Yon no real scale. Measured zero true positives across 46 corpus hits; every one was a genuine state transition.landscapeused figuratively. "Competitive landscape" is a term of art; "the landscape of modern tooling" is the tell.
Structural patterns
| Pattern | Fix |
| "serves as", "stands as", "boasts" | Use "is/are/has" |
| "It's not just X, it's Y" | Just say Y |
| Stacked qualifiers ("could potentially possibly") | One qualifier |
| Hollow gerunds: "highlighting/underscoring/emphasizing" | Cut or make main clause |
Formatting tells
| Tell | Fix |
| Em dash overuse | Comma, parentheses, or colon. The validator measures the rate per document; a threshold breach means the whole draft, not one sentence. |
| No contractions | Use "don't", "it's", "we've" |
| Title Case Headings | Sentence case |
| Boldface overuse | Bold genuine emphasis only |
| Uniform paragraph length | Vary naturally |
Over-formality substitutions
| Avoid | Use |
| "In order to" | "To" |
| "Due to the fact that" | "Because" |
| "At this point in time" | "Now" |
| "Prior to" / "Subsequent to" | "Before" / "After" |
| "With respect to" | "About" or "For" |
| "Has the ability to" | "Can" |
What human writing has
- Burstiness: dramatic variation, a 3-word sentence next to a 25-word sentence is the target, not mild variation
- "And", "But", "Or" as sentence starters are fine
- Specifics over abstractions: name the file, cite the number
- Opinions: take a position; acknowledge complexity or mixed feelings
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.