agentleFS
Calix-LCalix-L/awesome-latex-skills/latex-polish/SKILL.md

latex-polish

Skill182 starsChanged 42 days agoMITView on GitHub

Edit academic prose in LaTeX manuscripts or supplied text at light, moderate, or strict intensity. Improve grammar and clarity while preserving scientific claims, uncertainty, numbers, citations, and LaTeX structure.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. flonat/flonat-research★ 150same name

    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 repository4 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    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:

    No level authorizes changing findings, adding evidence, or making a paper "publication ready" by declaration.

    line 19
  2. Very specific:

    Keep modality and scope: can, may, suggests, and under these conditions may carry scientific meaning.

    line 27
  3. Distinctive:

    Do not add significance tests, p-values, citations, mechanisms, novelty, or broader-impact claims.

    line 30
  4. Distinctive:

    Leave verbatim, listings, minted, and inline code untouched.

    line 37
  5. Distinctive:

    Keep an already clear sentence.

    line 38
  6. Distinctive:

    Read supporting material only as needed:

    line 43
  7. Very specific:

    If new compilation errors arose, repair the text edit or use latex-rescue when available.

    line 70

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Scope and intensity

Edit the requested text, section, or manuscript. Infer scope from the request; ask only when a missing preference changes the work. Default to moderate.

LevelEditing scope
lightGrammar, spelling, articles, agreement, and necessary syntax fixes.
moderateLight edits plus clearer phrasing, concision, and sentence flow.
strictModerate edits plus stronger paragraph organization within the requested scope.

No level authorizes changing findings, adding evidence, or making a paper "publication ready" by declaration. Use the author's chosen English variant, terminology, and venue requirements when supplied.

Preserve meaning and source structure

  • Preserve numbers, units, uncertainty, causal direction, comparisons, named entities, and the distinction between relative percentages and percentage points.
  • Keep modality and scope: can, may, suggests, and under these conditions may carry scientific meaning. Removing them can turn a capability into an observed result or a qualified claim into a universal claim.
  • Do not add significance tests, p-values, citations, mechanisms, novelty, or broader-impact claims. Flag missing evidence for the author instead.
  • Preserve citation/reference/label keys, bibliography records, equations, macro definitions, file paths, comments, and quoted material. Treat LaTeX arguments by their role, not by a blanket regular-expression replacement.
  • Prose inside \emph{}, \textbf{}, headings, or captions may be edited when it falls within scope; retain the command, options, braces, keys, and math. Leave verbatim, listings, minted, and inline code untouched.
  • Keep an already clear sentence. Formality does not require heavier synonyms, and sentence length alone is not a reason to split it.

Edit with relevant guidance

Read supporting material only as needed:

  • Style guardrails: voice, clarity, notation, consistency, and claim strength. These are editing choices, not universal laws.
  • Section anatomy: section-specific purposes; adapt to the article type and the verified target requirements.
  • Chinglish patterns: common constructions when they actually occur; do not infer errors from the author's identity.
  • Academic phrasebank: optional phrasing patterns. Fill only with claims already supported by the manuscript.

Lead results prose with the finding and its actual evidence. Improve method descriptions without inventing missing reproducibility details. Suggest missing limitations or ethics content for author review; formatting requirements belong to the specified venue/year/track/stage, not to generic prose polishing.

Verify and deliver

Review the diff for accidental changes to meaning, scientific tokens, commands, and argument balance. For a local project, use its configured engine/backend to check compilation when available. For pasted text or missing build tools, mark compilation as not performed or unverified rather than claiming success.

Return the edited text or changed files, intensity, representative before/after edits with reasons, substantive questions for the author, and observed build/ visual-check status. Scale the explanation to the edit; a large manuscript can use a grouped change summary plus the full diff rather than an item for every word. If new compilation errors arose, repair the text edit or use latex-rescue when available. Preserve unrelated author changes during rollback.

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.