agentleFS
KangDongGwonKangDongGwon/claude-skills/skills/paper-sections/SKILL.md

paper-sections

Skill0 starsChanged 6 months agoMITView on GitHub

Draft IMRAD paper sections (Introduction, Methods, Results, Discussion, Outlook) in the author's voice, using STYLE_PROFILE.md templates. Includes numeric-superscript citation format rules. Forked from scientific-writing. Use when writing or revising a specific section of a scientific manuscript, or when inserting citations. Triggers - paper draft, IMRAD, introduction draft, results 작성, discussion 섹션, 섹션 초안, manuscript section, citation format, 인용 포맷, bibliography, reference format, in-text cite, superscript citation.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 9 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    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:

    Every sentence passive. Zero first person.

    line 24
  2. Distinctive:

    Consider bridge B3/B4 for multi-limitation synthesis, B5 for close.

    line 37
  3. Very specific:

    Start with In summary, ...

    line 40
  4. Distinctive:

    No hedging adverbs. No rhetorical close.

    line 41
  5. Distinctive:

    No space between word and superscript.

    line 53
  6. Distinctive:

    Never cite a review when a primary source exists for the specific claim.

    line 58
  7. Distinctive:

    When target journal unknown, emit EndNote-style block:

    line 63
  8. Distinctive:

    Never fabricate DOI, year, or author list.

    line 73
  9. Distinctive:

    If any required metadata is missing, return FLAG with the missing field name.

    line 74
  10. Distinctive:

    When inserting citations, use the format in section 7 above.

    line 83
  11. Very specific:

    No banned word (STYLE_PROFILE 3.1) may appear.

    line 88

≈ “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

paper-sections

Fork of global scientific-writing skill, specialized to the author's voice via paper-style. Absorbs former paper-citation-format skill as section 7 below.

Dependencies

  • Load paper-style Mode 1 first. Use its output as hard constraint.

Section playbooks

Introduction (target: 3-5 paragraphs)

  1. Topic paragraph - field-level fact opener. Define acronyms parenthetically. Citation density 0.9-1.3 refs/sentence. See STYLE_PROFILE G1.
  2. Gap/motivation paragraph - quantitative gap statement, one subordinate clause. See STYLE_PROFILE G2. Consider bridge B1 (Paradox-Pivot) or B3 (Enumerated-Walls) sparingly.
  3. Solution/approach paragraph - solution announcement. See STYLE_PROFILE G3.
  4. Summary of contributions - In this work, we ... with numbered contributions.

Methods (passive, purpose-forward)

  • Every sentence passive. Zero first person.
  • Order: materials -> synthesis -> characterization -> computational -> kinetic.
  • Parameters in parentheses, not footnotes.

Results (passive + purpose opener)

  • Each subsection opens with purpose sentence (see STYLE_PROFILE 4.2).
  • Figures referenced parenthetically (Fig. 2a). NEVER As shown in Fig..
  • One finding per paragraph. Close with mechanistic consequence.

Discussion (mixed voice, active for claims)

  • Open with contribution statement + retrace (see G5).
  • we permitted for active mechanistic claims.
  • Citation density drops to 0.2-0.4 refs/sentence.
  • Consider bridge B3/B4 for multi-limitation synthesis, B5 for close.

Outlook / Conclusion

  • Start with In summary, ... or By bridging ... (see STYLE_PROFILE 4.4).
  • No hedging adverbs. No rhetorical close.

7. Citation format (absorbed from paper-citation-format)

Default style

Numeric superscript, following Nature Commun / ACS Catal convention.

  • Single: text^1
  • Range: text^1-4
  • List: text^1,3,5
  • Mixed: text^1-3,7

No space between word and superscript. Superscripts attach to the concept, not the sentence terminator. Attach before punctuation when the cite scopes the clause, after punctuation when it scopes the sentence.

In-text cite rules (author voice via STYLE_PROFILE)

  • Intro: ~1 cite per sentence. Clusters of ^1-4 acceptable.
  • Discussion: cite only for specific comparative claims or contradictions to prior art. Mechanistic own-work statements stay uncited.
  • Never cite a review when a primary source exists for the specific claim.
  • Novelty claim requires contrast cite: Unlike previous work that ...^12, here we ....

Bibliography entry format

When target journal unknown, emit EndNote-style block:

[N] Author, A. B.; Author, C. D. Title. Journal Year, Vol, Pages. DOI.

If journal known:

  • Nature Commun: Author, A. B. et al. Title. Journal Vol, Pages (Year).
  • ACS: Author, A. B.; Author, C. D. Title. Journal Year, Vol, Pages.

Citation hard rules

  • Never fabricate DOI, year, or author list.
  • If any required metadata is missing, return FLAG with the missing field name. Do not guess.
  • et al. only when author count > 3 in short form journals.

Workflow when invoked

  1. Receive args: section_name, optional subsection_identifier, optional existing_text, optional data_pointers.
  2. Load STYLE_PROFILE via paper-style Mode 1.
  3. Gather minimal context: read only the referenced data files. Do not load the whole manuscript.
  4. Draft paragraph(s) following the section playbook above.
  5. When inserting citations, use the format in section 7 above.
  6. Self-check against STYLE_PROFILE section 12 quick checklist before returning.
  7. Return: draft text + list of gaps/assumptions + suggested next command.

Hard rules

  • No banned word (STYLE_PROFILE 3.1) may appear.
  • Sentence length: 16-22 words mean, 30 word max.
  • One subordinate clause per sentence max.
  • Active/passive ratio matches section norm.

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.