Simple History Voice
bonny/WordPress-Simple-History
Pär's writing voice and per-text-type rules for Simple History, on top of the humanizer skill.
Proofread a PR description held in a .md file (e.g. An agent skill from ROCm/aotriton.
Install Proofread
The automated check flagged lines worth reading first. See the safety section below.
$ npx skills add ROCm/aotriton --skill proofread -a claude-codeProject install by default; add -g for ~/.claude/skills/.
$ gh skill install ROCm/aotriton proofread --agent claude-codeProject scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).
$ git clone --depth 1 https://github.com/ROCm/aotriton.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/proofread .claude/skills/proofread && rm -rf skills-srcUse ~/.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/
Install the "proofread" agent skill from https://github.com/ROCm/aotriton/tree/main/.claude/skills/proofread into .claude/skills/proofread/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "proofread", then confirm the skill loads.Claude Code copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$skill-installer install https://github.com/ROCm/aotriton/tree/main/.claude/skills/proofreadType this inside Codex. $skill-installer <name> installs a curated skill from openai/skills. The installer writes to $CODEX_HOME/skills (default ~/.codex/skills). Restart Codex if the skill does not show up.
$ npx skills add ROCm/aotriton --skill proofread -a codexProject install goes to .agents/skills/; add -g for ~/.codex/skills/.
$ gh skill install ROCm/aotriton proofread --agent codexProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ROCm/aotriton.git skills-src && mkdir -p .agents/skills && cp -r skills-src/.claude/skills/proofread .agents/skills/proofread && rm -rf skills-srcUse ~/.agents/skills/ instead of .agents/skills for a personal install.
Codex skills documentation · loads skills from .agents/skills/
Install the "proofread" agent skill from https://github.com/ROCm/aotriton/tree/main/.claude/skills/proofread into .agents/skills/proofread/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "proofread", then confirm the skill loads.Codex copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add ROCm/aotriton --skill proofread -a cursorProject install goes to .agents/skills/; add -g for ~/.cursor/skills/.
$ gh skill install ROCm/aotriton proofread --agent cursorProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ROCm/aotriton.git skills-src && mkdir -p .cursor/skills && cp -r skills-src/.claude/skills/proofread .cursor/skills/proofread && rm -rf skills-srcUse ~/.cursor/skills/ instead of .cursor/skills for a personal install.
Cursor skills documentation · loads skills from .cursor/skills/, .agents/skills/, .claude/skills/, .codex/skills/
Install the "proofread" agent skill from https://github.com/ROCm/aotriton/tree/main/.claude/skills/proofread into .cursor/skills/proofread/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "proofread", then confirm the skill loads.Cursor copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gemini skills install https://github.com/ROCm/aotriton.git --path .claude/skills/proofread--scope user (default) or --scope workspace; --path is the subfolder of the repo that holds the skill; --consent skips the security confirmation prompt.
$ npx skills add ROCm/aotriton --skill proofread -a gemini-cliProject install goes to .agents/skills/; add -g for ~/.gemini/skills/.
$ gh skill install ROCm/aotriton proofread --agent gemini-cliProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ROCm/aotriton.git skills-src && mkdir -p .gemini/skills && cp -r skills-src/.claude/skills/proofread .gemini/skills/proofread && rm -rf skills-srcUse ~/.gemini/skills/ instead of .gemini/skills for a personal install, then run /skills reload.
Gemini CLI skills documentation · loads skills from .gemini/skills/, .agents/skills/
Install the "proofread" agent skill from https://github.com/ROCm/aotriton/tree/main/.claude/skills/proofread into .gemini/skills/proofread/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "proofread", then confirm the skill loads.Gemini CLI copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ gh skill install ROCm/aotriton proofreadInstalls for Copilot at project scope by default; add --scope user for a personal install. Preview a skill first with gh skill preview. Needs GitHub CLI 2.90.0 or later (public preview).
$ npx skills add ROCm/aotriton --skill proofread -a github-copilotProject install goes to .agents/skills/; add -g for ~/.copilot/skills/.
$ git clone --depth 1 https://github.com/ROCm/aotriton.git skills-src && mkdir -p .github/skills && cp -r skills-src/.claude/skills/proofread .github/skills/proofread && rm -rf skills-srcUse ~/.copilot/skills/ instead of .github/skills for a personal install. Commit .github/skills so cloud agent and code review can use it.
GitHub Copilot skills documentation · loads skills from .github/skills/, .claude/skills/, .agents/skills/
Install the "proofread" agent skill from https://github.com/ROCm/aotriton/tree/main/.claude/skills/proofread into .github/skills/proofread/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "proofread", then confirm the skill loads.GitHub Copilot copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
$ npx skills add ROCm/aotriton --skill proofread -a opencodeOpenCode documents no install command of its own. Project install goes to .agents/skills/; add -g for ~/.config/opencode/skills/.
$ gh skill install ROCm/aotriton proofread --agent opencodeProject scope by default (.agents/skills/); add --scope user for a personal install.
$ git clone --depth 1 https://github.com/ROCm/aotriton.git skills-src && mkdir -p .opencode/skills && cp -r skills-src/.claude/skills/proofread .opencode/skills/proofread && rm -rf skills-srcUse ~/.config/opencode/skills/ instead of .opencode/skills for a personal install.
OpenCode skills documentation · loads skills from .opencode/skills/, .claude/skills/, .agents/skills/
Install the "proofread" agent skill from https://github.com/ROCm/aotriton/tree/main/.claude/skills/proofread into .opencode/skills/proofread/ in this project. Copy the whole folder (SKILL.md and every file beside it), keep the folder name "proofread", then confirm the skill loads.OpenCode copies the folder itself, the same result as the manual copy. Check what it changed before you commit it.
Facts
- Skill name
proofread- GitHub stars
- 101
- Token cost
- ~1k tokens
- SKILL.md length
- 530 words
- Files
- 1
- Skills in repo
- 1
- Repo updated
- First seen
- Licence
- MIT
At a glance
Proofread a PR description held in a .md file (e.g. An agent skill from ROCm/aotriton.
- Works in 2 steps: Fix in place, then list briefly → Report last — do not fix unilaterally
- Asked to proofread
- SKILL.md covers Reply order — least important…, 1. Fix in place, then list…, 2. Report last — do not fix… and Verify before flagging a…, plus 2 more sections
- Calls git
What it does
Proofread is an agent skill from ROCm/aotriton. Proofread a PR description held in a .md file (e.g. pr237.md, usually untracked). Fixes typos, grammar and formatting in place and lists them first, then reports content-level problems last — claims the code contradicts, inverted explanations, misclassified sections. Use whenever asked to proofread, review or check over a PR description, PR draft, or similar .md write-up.
Its SKILL.md is about 1k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.
It sits in Writing & Content, covering Copy editing and proofreading and Pull requests. The repository describes itself as: Ahead of Time (AOT) Triton Math Library. The licence is MIT.
When your agent uses it
- Asked to proofread
- Check over a PR description
- Similar .md write-up
Example prompts
- “/proofread”
Workflow steps
2 steps, taken from the step headings in SKILL.md.
What it can do on your machine
Read from SKILL.md and the folder at commit b5e8cfb. 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
Shell commands in SKILL.md call:
git
From the folder's file list and the shell code blocks in SKILL.md.
Network
No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.
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
Proofread loads about 1k tokens when it runs. Until then it costs about 96 tokens; SKILL.md has 530 words of instructions outside code blocks.
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: warnings
The automated check found patterns that need a careful read before installing.
- WarningTells the agent its actions are pre-authorized / not to stop for confirmationSKILL.md:27
Apply directly with `Edit` — never ask permission for these. Then list them
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 ROCm/aotriton at commit b5e8cfb, republished under its MIT licence (© ROCm). 530 words, ~1,019 tokens.
Download SKILL.mdSave it as
.claude/skills/proofread/SKILL.md (or your agent's skills folder).- name
- proofread
- description
- Proofread a PR description held in a .md file (e.g. pr237.md, usually untracked). Fixes typos, grammar and formatting in place and lists them first, then reports content-level problems last — claims the code contradicts, inverted explanations, misclassified sections. Use whenever asked to proofread, review or check over a PR description, PR draft, or similar .md write-up.
Proofreading a PR description
The author wants a clean file and a reply they can read from the bottom up.
Reply order — least important FIRST
The author reads the tail of a long reply first. So the reply runs in ascending order of importance: the mechanical fixes open it, the content problems close it, and the single most serious finding is the last thing on the page. This inverts the usual habit of leading with the headline — do not "fix" it back.
1. Fixed in place <- mechanical, already applied, skimmable
2. Content findings <- ascending severity, most serious LASTIf there are no content findings, say so in one line at the end.
1. Fix in place, then list briefly
Apply directly with Edit — never ask permission for these. Then list them
at the TOP of the reply, compressed: a one-line summary plus counts beats an
itemised inventory ("3 typos, 2 missing articles, reflowed 4 lines >78ch").
Name individual fixes only where the author might disagree with the choice.
- Typos, misspellings, grammar, missing or wrong articles
- Wrapping: reflow to 78 characters (
awk 'length>78' file.mdto find them) - Punctuation consistency within the document — pick whichever convention the majority of bullets already use and make the rest match
- Stray double blank lines, trailing whitespace, inconsistent list markers
- Inconsistent backticking or capitalisation of identifiers that appear elsewhere in the same file
2. Report last — do not fix unilaterally
State these at the END of the reply, with the correction you would make. Ascending severity, so the worst one lands last. Keep each one short.
Listed here in the same ascending order the reply should use:
- Budget overruns per prwriter's line/word ceilings, when substantial.
- Unsupported numbers. House style wants the command or measurement behind any perf/scale claim.
- Misclassification against
.claude/agents/prwriter.md: a headline behaviour change buried in Minor Changes, a real caveat with no Known Issues entry, sections out of canonical order. - Technical explanations that are wrong or misleading even where not strictly false, e.g. naming the wrong mechanism for a real symptom.
- Claims the code contradicts. The most valuable thing this skill does, so it goes last where the author reads first. A description can confidently assert the opposite of what the diff does — an inverted flag polarity, a mode claimed not to apply when it does, a symbol that was never touched.
Show full SKILL.md (158 more words)Show less
Verify before flagging a content error
Never flag a technical claim from memory or from earlier conversation. Read the source or the diff first:
git diff <base>..HEAD # what the PR actually changes
grep -n '<symbol>' <file> # what the symbol actually doesQuote the file:line you checked in the reply. A wrong correction costs more than a missed typo.
Respect the author's edits
- The file changes between turns. Re-read it immediately before editing — the author frequently rewrites sections after asking.
- If the author deleted a section, it was deliberate. Do not restore it; note its absence once at most, then leave it.
- If the author overrules a finding, drop it and do not re-raise it. Their
common and correct objection: "the diff makes this obvious" — prwriter
says to omit detail that is self-describing from the code, so a one-line
mechanical fix genuinely does not need a
why.
Housekeeping
These files are usually untracked scratch (pr<N>.md at the repo root).
Do not git add them, do not commit them, and never add them to
.gitignore — see CLAUDE.md.
© ROCm, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file
Files
Just SKILL.md in .claude/skills/proofread of ROCm/aotriton.
Open the folder on GitHubat commit b5e8cfb
Compare with similar skills
Proofread 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.
| Skill | Stars | Used in | Tokens | Auto-check | Licence | Repo updated |
|---|---|---|---|---|---|---|
| Proofread this skillROCm/aotriton | 101 | — | ~1k | Automated safety check: Warn | MIT | |
| Simple History Voicebonny/WordPress-Simple-History | 317 | — | ~775 | Automated safety check: Pass | None | |
| Tech Writerstacklok/mecatl | 241 | — | ~2.2k | Automated safety check: Pass | Apache-2.0 | |
| User-Facing Text Cleanupguillaumemeyer/watermarks-remover | 24k | — | ~3.5k | Automated safety check: Pass | MIT | |
| Story Multi-Perspective Reviewzenstory-ai/oh-story-claudecode | 7.4k | 3 repos | ~3k | Automated safety check: Pass | MIT | |
| Simple Issue Descriptionevery-app/open-seo | 23k | 1 repos | ~1.2k | Automated safety check: Pass | MIT |
Similar skills
Tech Writer
stacklok/mecatl
A skill your agent uses when writing or substantively editing user-facing documentation: drafting a new page, rewriting or restructuring an existing one, adding a major section, or turning…
User-Facing Text Cleanup
guillaumemeyer/watermarks-remover
Audits prose for invisible Unicode characters and rewrites it while keeping facts, citations, code and required disclosures unchanged and the writer's voice intact.
Story Multi-Perspective Review
zenstory-ai/oh-story-claudecode
Reviews Chinese web-novel text with several reviewer agents in parallel, falling back to a single-agent pass, and reports structure, character, prose and setting problems with fixes.
Simple Issue Description
every-app/open-seo
Turn a rough bug report, feature request, support note, or pull request into a short, plain-language issue focused on the problem and desired behavior.
Chinese Text Humanizer
op7418/Humanizer-zh
Edits Chinese articles, comments and documents to remove filler, repetition and template phrasing while keeping the facts, the level of certainty and the author's voice.
Related
Categories
Questions about Proofread
What does Proofread do?
Proofread a PR description held in a .md file (e.g. An agent skill from ROCm/aotriton. Proofread is an agent skill from ROCm/aotriton.g.
When should I use Proofread?
Proofread fits situations like: asked to proofread; check over a PR description; similar .md write-up.
How do I install Proofread in Claude Code?
Run `npx skills add ROCm/aotriton --skill proofread -a claude-code`. Or copy the skill folder (.claude/skills/proofread in ROCm/aotriton) into .claude/skills/proofread in your project. Claude Code loads it when a task matches its description.
How do I install Proofread in Codex?
Run `npx skills add ROCm/aotriton --skill proofread -a codex`. Or copy the skill folder (.claude/skills/proofread in ROCm/aotriton) into .agents/skills/proofread in your project. Codex loads it when a task matches its description.
Can I use Proofread 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 ROCm/aotriton --skill proofread -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/proofread, .gemini/skills/proofread, .github/skills/proofread and .opencode/skills/proofread in your project.
What does Proofread need to run?
Going by SKILL.md and its folder, Proofread needs the command-line tools its instructions call (git).
Does Proofread access the network?
SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.
Is Proofread safe to install?
Our automated static check of SKILL.md flagged 1 warning(s): tells the agent its actions are pre-authorized / not to stop for confirmation. Read the flagged lines before installing; the check is not a guarantee either way.
What licence does Proofread use?
Proofread 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 Proofread use?
About 1k tokens (SKILL.md is roughly 4.1k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.
What are the alternatives to Proofread?
Skills that share tags, products or a category with Proofread: Simple History Voice (bonny/WordPress-Simple-History, 317 stars), Tech Writer (stacklok/mecatl, 241 stars), User-Facing Text Cleanup (guillaumemeyer/watermarks-remover, 24k stars) and Story Multi-Perspective Review (zenstory-ai/oh-story-claudecode, 7.4k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.
Who maintains Proofread?
ROCm (a GitHub organization) maintains it in ROCm/aotriton, which has 101 GitHub stars. The repository was last updated on October 6, 2026.
Source: ROCm/aotriton on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.