academic-figure-workflow
Skill8 starsChanged 5 months agoMITView on GitHub
Create, revise, or inspect evidence-backed academic plots and diagrams with editable sources. Use for figure work itself, not manuscript prose or page formatting. 适用于论文作图、修图与图形质量检查。
1 other project uses a version of this.
Adopted by
Most recently taken up by
- Saddss/cursor-skillstheir own version, first seen Aug 29, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- Saddss/cursor-skills★ 2same 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 repository9 other files it gives its agents.
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 28
Read the relevant branch, not this entire reference collection.
- Distinctive: line 73
Do not reset the recorded review count to obtain approval.
≈ “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.
Files
Academic Figure Workflow
Preserve scientific meaning and the user's requested scope. Do not invent data, mechanisms, code structure, or experimental images. Retain editable sources and notation; a screenshot is not a replacement for a figure source.
Choose The Task Mode
| Request | Work and completion |
| Local edit or read-only inspection | Work with the existing SVG, Draw.io, or plotting source. Preserve user styling; do not redraw or migrate to YAML unless requested. Inspect affected rendering/exports, report the change or findings, and do not claim whole-figure publication approval from a local check. Read-only inspection does not modify the source. |
| New draft or design exploration | Use supplied facts and clearly identified assumptions. Choose an appropriate editable format; no full manuscript intake is needed. Missing evidence limits claims and publication status, not the ability to make a labelled concept draft or storyboard. |
| Reproducible or submission package | Use the bundled YAML generator and package workflow. Preserve its provenance, editable-source, export, and QA contract. Draft package generation is available without claiming submission readiness. |
Choose from the requested deliverable; ask only when an unresolved choice materially changes the result. A simple edit does not require claim anchors, new sidecars, or full-package QA unless it modifies an existing package whose consistency must be maintained. In that case update its canonical source and rebuild affected outputs; do not leave a stale manifest or approval.
Choose Relevant Guidance
Read the relevant branch, not this entire reference collection.
| Decision or task | Reference |
| Chart selection, data mappings, or statistical encodings | Data visualization advisor, before plotting |
| Complex, multi-panel, or evidence-sensitive figure argument | Figure argument contract |
| New SVG mechanism or Draw.io engineering diagram | SVG and Draw.io diagrams |
| Authoring a generator spec | FigureSpec schema; for diagrams, diagram spec |
| Model architecture layout | Architecture contract |
| Dense workflow layout or engine comparison | Flowchart engine comparison |
| Microscopy, photographs, or other image panels | Image integrity, before image processing |
| AI-generated supporting bitmap | AI image-generation gate, before generation |
| Choosing between production methods | Method selection |
| Final submission inspection without a package | Rendered figure QA |
| Publication package acceptance | Package QA |
| Design examples or an unfamiliar figure type | Reference gallery; obey its asset licence and private-reference restrictions |
For new work, prefer Matplotlib for quantitative plots, Draw.io for editable
workflows/engineering diagrams, and SVG for precise mechanism figures. Respect
an existing source or explicitly chosen tool. Office/PPT/WPS support in the
bundled generator is capability detection only, not native editing; use
python scripts/figure_package.py capabilities when that capability matters.
Defaults And Exceptions
Scientific integrity and authorization boundaries always apply. Within those boundaries, use the requested purpose and target venue/template first, then the matching profile, then defaults for unspecified details. If the request conflicts with the target submission guide, resolve that conflict rather than silently choosing one or claiming compliance.
For submission styling, consult publication standards. For format, resolution, or font requirements, consult export standards. For colour/encoding decisions, consult accessibility standards. These are conditional references, not a prerequisite for every edit. Nature-like dimensions and external-title conventions are defaults for that use, not rules for all figures. Do not waive a failing package check to claim a venue override; if the generator cannot express a requirement, report the limitation and retain a draft or use a suitable authorized source/export workflow.
Finish The Requested Work
Continue source corrections and relevant checks while they make progress. Repeated failure without progress calls for a different layout/chart strategy, simplification within scope, or a concrete blocker—not a fixed pass-count cap. Do not reset the recorded review count to obtain approval.
Return the requested source/export or inspection findings, relevant validation status, and unresolved issues. A publication claim requires actual final-size visual inspection and applicable export/evidence checks; command success alone is insufficient. Do not expand a local edit into a manuscript audit or return an unfinished first implementation when the requested checks remain feasible.
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
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.