agentleFS
jjfroehlichjjfroehlich/agent-skills-for-academic-research/skills/scientific-writing/SKILL.md

scientific-writing

Skill2 starsChanged 30 days agoMITView on GitHub

Draft or revise manuscripts, abstracts, figure legends, thesis chapters, and scientific arguments for readers. Exclude technical documentation, grant writing, and publication-process strategy.

44 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use usfiscaldata, research-grants and 10 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.
  • yf8578Skill

    usfiscaldata

    Query the U.S. Treasury Fiscal Data API for federal financial data including national debt, government spending, revenue, interest rates, exchange rates, and savings bonds. Access 54 datasets and 182 data tables with no API key required. Use when working with U.S. federal fiscal data, national debt tracking (Debt to the Penny), Daily Treasury Statements, Monthly Treasury Statements, Treasury securities auctions, interest rates on Treasury securities, foreign exchange rates, savings bonds, or any U.S. government financial statistics.

    Also in 7 projects that use this

  • garri333Skill

    research-grants

    Write competitive research proposals for NSF, NIH, DOE, DARPA, and Taiwan NSTC. Agency-specific formatting, review criteria, budget preparation, broader impacts, significance statements, innovation narratives, and compliance with submission requirements.

    Also in 10 projects that use this

  • garri333Skill

    arboreto

    Infer gene regulatory networks (GRNs) from gene expression data using scalable algorithms (GRNBoost2, GENIE3). Use when analyzing transcriptomics data (bulk RNA-seq, single-cell RNA-seq) to identify transcription factor-target gene relationships and regulatory interactions. Supports distributed computation for large-scale datasets.

    Also in 9 projects that use this

  • Hongyu-yuSkill

    torch-geometric

    Graph Neural Networks (PyG). Node/graph classification, link prediction, GCN, GAT, GraphSAGE, heterogeneous graphs, molecular property prediction, for geometric deep learning.

    Also in 9 projects that use this

  • SoljournerSkill

    get-available-resources

    Enumerate installed packages, databases, and tools for engineering workflows

    Also in 9 projects that use this

  • +7 more from garri333

The file

Files
SKILL.md

Scientific Writing

Purpose

Help draft, diagnose, and revise scientific writing so the reader can see the claim, evidence, caveats, and section logic without losing scientific accuracy.

Use this skill when

  • First classify the requested artifact and output: continue only when audience-facing scientific prose or argument is primary, not merely because a technical document discusses science.
  • The user provides scientific prose and asks for a rewrite, critique, outline, or readiness check.
  • The artifact is an abstract, introduction, results paragraph, discussion, conclusion, figure legend, thesis chapter, author contribution statement, or acknowledgement.
  • The user asks for clearer flow, a sharper gap or claim, less overclaiming, better section structure, or manuscript-ready wording.
  • The request needs a concrete edited artifact, not only general advice.

Do not use this skill when

  • The request concerns computational or analysis plans, pipeline or implementation documentation, README files, code comments, object names, variable names, or terminology inside scripts without a primary audience-facing scientific argument.
  • The request is mainly journal selection, submission logistics, peer-review strategy, grant writing, slide design, poster layout, or literature synthesis without a writing deliverable.
  • The user asks for statistical analysis, code, figure generation, or experimental design rather than prose revision.
  • Authorship questions require institutional, legal, or policy determinations beyond wording and documentation guidance.

Core workflow

  1. Match the intervention to the request. For local wording or compression, edit from the supplied prose and constraints; for structural review, establish the artifact, audience, claim, and evidence boundary. Ask only for missing context that changes the work.
  2. Diagnose section or paragraph function when the problem is structural; do not require a structural diagnosis for a settled local edit.
  3. Decide the intervention level before rewriting:
    • Structure: missing claim, wrong section job, chronology instead of reader logic, overloaded paragraph, or weak transition.
    • Rhetoric: missing gap, result, implication, limitation, comparison, or contribution.
    • Sentence: hidden actor, noun stack, vague verb, unstable term, overlong modifier chain, or overclaim.
  4. For whole-manuscript or thesis-chapter review, trace the question, evidence, finding, conclusion, and limits. For an excerpt, check only the claims it makes.
  5. Consult a reference or checklist when its section-specific guidance would improve the requested revision.
  6. Preserve meaning, uncertainty, statistical direction, actor, scope, caveats, and terminology while tightening language. Add facts only from supplied evidence; flag consequential omissions or use explicit placeholders. A smoother story cannot replace a missing control or unsupported inference.
  7. Return a concrete output: revised text, section outline, prioritized revision plan, checklist review, or before/after diagnosis.

Output formats

Select only the format that fits the requested deliverable; these are patterns, not mandatory response sections.

  • Diagnosis + rewrite: brief problem statement, revised prose, and rationale for key changes.
  • Section plan: ordered rhetorical moves with one-sentence job for each paragraph or subsection.
  • Readiness review: prioritized findings, required fixes and unresolved checks with their evidence limits.
  • Compression pass: what to cut, combine, or move while preserving the claim.
  • Contribution wording: careful authorship, contribution, or acknowledgement language with caveats.

Reference routing

  • Open references/abstracts.md for abstracts, summaries, graphical-abstract text, or compressed paper descriptions.
  • Open references/introductions.md for gap, significance, aim, scope, or field-context problems.
  • Open references/results-sections.md for results paragraphs, figure callouts, local interpretation, and format-dependent placement of broader claims.
  • Open references/discussions.md for interpretation, limitations, implications, conclusions, and overclaim control.
  • Open references/figure-legends.md for figure captions, panel logic, labels, statistics, and self-contained legends.
  • Open references/paper-structure.md for whole-manuscript flow, Methods reporting, central contribution, titles/headings, paragraph order and sentence-level style choices.
  • Open references/thesis-writing.md for thesis/dissertation chapters and institution-sensitive structure.
  • Open references/authorship-and-contributions.md for contribution statements, acknowledgements, or authorship-sensitive wording.
  • Open references/handbook-access.md when a fuller explanation or worked example would help with the current task.

Quick checklist

Apply only the checks relevant to the requested scope and artifact. Inspect supplied context first; treat identify or clarify as analysis when the answer is already available, and ask only for missing information that could change the result.

  • What is the one main contribution or question?
  • Does the section do its specific job, or is it borrowing another section's job?
  • Can a reader identify the gap, method or approach, evidence, interpretation, and caveats?
  • Can each main result be traced to a question, figure/table, comparison, and bounded local conclusion?
  • Are claims calibrated to the supplied evidence?
  • Does the order explain the argument, including chronology where sequence itself matters?
  • Are important terms searchable and stable, especially in abstracts, titles, figure legends, and contribution statements?
  • Does the rewrite keep the author's scientific meaning, comparison, quantity, actor, scope, and uncertainty while reducing friction?

Common pitfalls

  • Polishing vague prose before fixing the argument.
  • Using research chronology when it hides the evidential logic, or deleting a sequence that explains it.
  • Removing necessary local interpretation from Results, or repeating Results without synthesis in Discussion.
  • Replacing accurate uncertainty with confident overstatement.
  • Cutting searchable method, system, comparison, or result terms during compression.
  • Treating acknowledgements as hidden authorship credit or contribution statements as praise.
  • Applying abstract or introduction patterns to every section.

Quality bar

The answer should be section-specific, concrete, and conservative with evidence. It should improve reader orientation, preserve uncertainty, avoid generic writing advice, and return usable prose or a prioritized plan unless the user asked only for diagnosis.

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.