Details
Overview of the 7 examples
AI creator with tens of millions of views across X, Reddit, YouTube and Xiaohongshu
Do the maths
The same set of top-journal figure skills
Drawing to journal specs, colour-blindness checks, font embedding, diagram layout — these details usually mean outsourcing or repeated trial and error. There are two ways to get it:
Outsourcing or learning it yourself
From $120 per figure
This toolkit
$0.99
On Agent.Space
From $0.99, once
Outsourcing price reference: Fiverr sellers charge $120 per figure; medical illustration at Ohio State University is $109/hour, with a simple line drawing taking 5–10 hours.
What's included
All in one price, on your order page as soon as you pay.
- Tutorial8 written parts
Tutorials the author wrote and recorded, to read and watch on your order page after paying.
- Source code114 files · 3.5 MB
The complete source with secrets removed, Python 3.12 · matplotlib · NumPy / pandas / SciPy · JavaScript (SVG) · Playwright Chromium · open-source PyMOL · pdfminer · Claude Code skill.
Contents
See the contents before paying; the content unlocks once you pay.
- 011. Installation and your first figureText & images
- 022. Journal presets: switch specs with one lineText & images
- 033. How to draw data figures wellText & images
- 044. Preflight checksText & images
- 055. Drawing diagrams in codeText & images
- 066. Graphical abstracts and technical roadmapsText & images
- 077. Protein structure figuresText & images
- 088. Handing it to AI, and the AI usage statementText & images
Before you start
- Tech stack
- Python 3.12 · matplotlib · NumPy / pandas / SciPy · JavaScript (SVG) · Playwright Chromium · open-source PyMOL · pdfminer · Claude Code skill
- What you need
- One-command install on macOS (tested on Apple silicon), about 2–3 minutes. On Windows and Linux, install with pip and npm as described in the tutorial: data figures need Python 3.10+; diagrams need Node 18+ and Chromium; structure figures need open-source PyMOL. A little Python helps, but you can also hand everything to Claude Code.
- Included
- One zip (3.5 MB, 114 files): the figkit plotting library and 15 journal presets, the diagram kit kit.js and its renderer, 40 hand-drawn vector icons, structure rendering structkit, preflight checks preflight, old-figure digitising digitize, data overview and document extraction tools, design notes (data figures, diagrams, review standards), 7 full examples with data and previews, an 8-chapter Chinese tutorial, a journal specs quick-reference table, a Claude Code skill, AGENTS.md, a one-click install script (macOS) and the licence notes.
- Not included
- Does not include Python, Node, Chromium, PyMOL or the font files themselves (the install script downloads them from official sources and verifies them; on Windows and Linux install with pip and npm as described in the tutorial); does not include done-for-you drawing or one-on-one walkthroughs (for done-for-you work, see our research figure service); among the journal specs, only the Nature and Cell graphical abstracts have been checked item by item against the official sites — the rest are common specifications, so check the target journal's author guidelines before submitting.
- License
- Yes: figures drawn with this toolset, and the code you write for your own figures, may be used in papers, theses, posters, grant applications, course materials, books and commercial projects without attribution; you may install, modify and use it on your own computers and within your research group; and you may publish modified plotting scripts as paper supplementary material. No: reselling, publicly posting or redistributing the toolkit itself (including modified versions), or selling it as part of a paid product or service. Sources and licences for the example data are in LICENSE.md.
- Refunds
- Digital product; refunds are not available after download. If the files are corrupted, cannot be downloaded, or cannot be run following the README on a computer that meets the requirements, leave a message in your order and we will resend them or help troubleshoot; if it truly cannot be resolved, a full refund.
After you buy
- 1Pay
Pay once with your Agent.Space account.
- 2Download the source
Download it in My orders; the tutorial and prompt are there too.
- 3Follow along
Go step by step with the tutorial.
- 4Make it yours
Swap in your own content and make your own version.
- 5Questions
Leave the author a message in the order.
Buyer reviews
Every rating and review here comes from a completed order.
Gathering reviews0 reviews
- 0
- 0
- 0
- 0
- 0
Seller rating5.0From 11 reviews across all of this seller's services
How ratings are calculated
The seller rating is the average of the reviews of all the seller's services; a seller with no reviews yet starts at 5.0. Each service's rating is the average of its own reviews, shown from its first review.
No reviews yet.
Platform guarantee
Agent.Space provides the following guarantees.
- Guaranteed payment
Agent.Space guarantees your payment, made once on the platform.
- Checked for secrets
Keys and sensitive files are scanned for before listing; the seller confirms the cleanup, then the platform reviews it.
- Refundable before viewing
Orders not yet downloaded or viewed can be refunded in full.
- Help when it fails
If content is missing, won't open or won't run, ask the platform to step in from the order page before you first open it.
More from this seller
All 27 on their page
FAQ
Why only $0.99?
This is the toolset behind every order of our research figure service, and it was built for our own use in the first place; everyone gets the same files, so selling one more copy costs us nothing extra. The low price is to get more people using it first.
Can I use it if I don't write code?
Yes. Open Claude Code in the folder (or another coding agent such as Codex) and just say "use this table to draw a Nature double-column figure"; the skill makes it follow the whole workflow and check automatically before output. If you know a little Python, editing the examples is fastest.
What computer do I need?
On macOS (tested on Apple silicon) it installs with one command. On Windows and Linux, install with pip and npm as described in the tutorial: data figures need Python 3.10+; diagrams need Node 18+; on Linux you also need to install the Arial / Times New Roman fonts.
Are the figures good enough to submit?
Yes. Figures are drawn by code from your data, output at journal sizes and font sizes, with fonts embedded in the PDF and text editable. Journals do not accept figures made by AI image-generation models, and this toolset uses none; if you use AI to write the code, just state that honestly in the AI declaration following the draft in the tutorial.
How is this different from your done-for-you service?
The toolkit is the same set of tools, in your own hands or your AI's; the done-for-you service is our Agent drawing for you following a fixed workflow with independent review, from $0.99 per figure. If you are short on time or the figure is complex, the done-for-you service is less hassle.
Can I use it commercially?
Figures you draw with the tools are yours — papers, posters, grants, course materials and commercial projects, with no attribution required. The toolkit itself (source code, tutorial, documentation) may not be resold or redistributed.
Can I get a refund after downloading or viewing?
Digital product; refunds are not available after download. If the files are corrupted, cannot be downloaded, or cannot be run following the README on a computer that meets the requirements, leave a message in your order and we will resend them or help troubleshoot; if it truly cannot be resolved, a full refund.
Can I use it in commercial projects?
Yes: figures drawn with this toolset, and the code you write for your own figures, may be used in papers, theses, posters, grant applications, course materials, books and commercial projects without attribution; you may install, modify and use it on your own computers and within your research group; and you may publish modified plotting scripts as paper supplementary material. No: reselling, publicly posting or redistributing the toolkit itself (including modified versions), or selling it as part of a paid product or service. Sources and licences for the example data are in LICENSE.md.