agentleFS
Robertmorrisluminousenergy177Robertmorrisluminousenergy177/paper-deep-reading-skill/paper-deep-reading/SKILL.md

paper-deep-reading

Skill1 starsChanged 4 months agoMITView on GitHub

Create beginner-friendly deep-reading notes for journal and research paper PDFs. Use when Codex needs to learn a paper, translate it paragraph by paragraph, build an Obsidian knowledge-base note, extract and place figures, write detailed figure notes, explain concepts for beginners, maintain a question log, or export the note to PDF. Also use for Chinese requests such as 学习文献, 精读论文, 逐段翻译, 图注解释, 问题荟萃, and Obsidian 笔记.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. Xuzhen-Li/literature-review-skills★ 2same name
    2. Jinsong-Zhou/bio-research-skills★ 0same 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.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 10 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈10

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use the PDF skill or an equivalent render-and-inspect workflow whenever the source is a PDF.

    line 12
  2. Very specific:

    Create a vault-style output folder with assets/figures/, exports/, and one main Markdown note unless the user asks for multiple notes.

    line 17
  3. Distinctive:

    Add beginner support at first use of key concepts, methods, axes, formulas, materials, and mechanisms.

    line 21
  4. Distinctive:

    Verify Markdown links, figure paths, section coverage, and whether the output is readable in Obsidian.

    line 23
  5. Distinctive:

    Read only the reference files needed for the user's request:

    line 28
  6. Distinctive:

    Prefer one substantial Markdown note over many small files.

    line 48
  7. Distinctive:

    Keep images close to the text that explains them.

    line 49
  8. Distinctive:

    Write figure notes specific to the paper, not generic captions.

    line 50
  9. Distinctive:

    Avoid unsupported claims. Say when a point is an inference from the paper.

    line 52
  10. Distinctive:

    Do not publicly include copyrighted full-text translations in example repositories unless the source license permits it.

    line 53
  11. Distinctive:

    Before finishing, report the note path, figure count, question-log status, and any export path.

    line 54

≈ “Common” is an estimate: at least 1 in 5 of the 372,429 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 Deep Reading

Purpose

Turn a journal paper into a polished learning artifact: one main Obsidian Markdown note with paragraph-level translation, placed figures, detailed figure notes, beginner explanations, glossary support, a question log, and optional PDF export.

Use the PDF skill or an equivalent render-and-inspect workflow whenever the source is a PDF. For two-column journal articles, rendered pages are the source of truth for reading order and figure placement.

Default Workflow

  1. Inspect the source PDF: title, authors, sections, page count, figures, captions, tables, references to supplements, and any copyright-sensitive constraints.
  2. Create a vault-style output folder with assets/figures/, exports/, and one main Markdown note unless the user asks for multiple notes.
  3. Translate and explain paragraph by paragraph. Number natural paragraphs as P01, P02, etc.; correct OCR or extraction order against page renders.
  4. Insert each figure near the paragraph that first uses it. Store images under assets/figures/ with stable descriptive names.
  5. For every main figure, write a figure note that explains what type of figure it is, what each panel does, what claim it supports, why it matters, and why the authors included it.
  6. Add beginner support at first use of key concepts, methods, axes, formulas, materials, and mechanisms. Mark claim boundaries: measured, calculated, inferred, or proposed.
  7. Maintain a question log for knowledge questions. Merge repeated questions into the closest existing entry instead of duplicating long explanations.
  8. Verify Markdown links, figure paths, section coverage, and whether the output is readable in Obsidian.
  9. Export to PDF only when requested. Render the note, inspect representative pages visually, and place final PDFs in exports/.

Required References

Read only the reference files needed for the user's request:

  • references/reading-workflow.md: full paper-to-note workflow and quality gates.
  • references/figure-note-guide.md: how to crop, place, and explain figures.
  • references/beginner-explanation-guide.md: how to explain a difficult field to a novice without oversimplifying.
  • references/obsidian-note-style.md: note structure, callouts, links, and CSS expectations.
  • references/question-log-guide.md: how to maintain the learning question log and merge repeated concepts.

Bundled Resources

  • assets/note-template.md: starting structure for the main paper note.
  • assets/question-log-template.md: structure for the separate question log.
  • assets/figure-note-template.md: reusable figure-note block.
  • assets/obsidian-css-snippet.css: optional Obsidian CSS for readable callouts and figures.
  • scripts/validate_note_structure.py: lightweight validation for headings, images, callouts, and repeated question IDs.
  • scripts/export_obsidian_note_pdf.js: Markdown/Obsidian image rendering to PDF through Playwright.
  • scripts/extract_pdf_assets.py: optional PyMuPDF helper to render pages and extract embedded images.

Output Standards

  • Prefer one substantial Markdown note over many small files.
  • Keep images close to the text that explains them.
  • Write figure notes specific to the paper, not generic captions.
  • Explain graph axes before interpreting trends.
  • Avoid unsupported claims. Say when a point is an inference from the paper.
  • Do not publicly include copyrighted full-text translations in example repositories unless the source license permits it.
  • Before finishing, report the note path, figure count, question-log status, and any export path.

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.