synthetic-sciences/openscience is an AI workbench that carries out scientific research by reading papers, forming hypotheses, writing and running code, conducting experiments, analyzing results, and preparing reports. Researchers use it for work in machine learning, biology, physics, and chemistry with remote or local models. Catalogue add-ons extend its scientific workflows through skills and instructions.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add synthetic-sciences/openscience --skill figuresgit clone --depth 1 https://github.com/synthetic-sciences/openscienceWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/synthetic-sciences/openscience/figures)<a href="https://agentmods.dev/skills/synthetic-sciences/openscience/figures"><img src="https://agentmods.dev/badge/skills/synthetic-sciences/openscience/figures/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/synthetic-sciences/openscience/figures"><img src="https://agentmods.dev/badge/skills/synthetic-sciences/openscience/figures.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00099 | $0.01868 |
| Opus 5.5 | $0.00040 | $0.00747 |
| Sonnet 5.5 | $0.00020 | $0.00374 |
| Haiku 4.5 | $0.00010 | $0.00187 |
Grade A, and why
figures scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 22d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 142 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Figures
A figure in a paper is an argument, not a screenshot of an array. It makes one claim, and a reader who skips the prose should still get that claim right. Default plotting output does not clear that bar: it is sized for a screen, titled where a caption belongs, colored from a cycle that collapses in greyscale, and rasterized where the document wants vector.
Non-negotiables
- Build at the final printed size. Pick the width from the document (
COLUMN3.25 in,TEXT5.5 in,WIDE6.75 in in the style module) and include it withwidth=\linewidth. Never build big and rescale: a 15-inch canvas dropped into a 5.5-inch column turns 11 pt tick labels into 4 pt. - Vector out. PDF for
\includegraphics, SVG beside it for preview. A rasterized plot blurs under the zoom every reviewer uses. PNG is only for genuinely raster content: a photograph, a sample grid, an attention map at pixel resolution. - Every number comes from a run. Read tracked metrics with the experiments tool
(
experiments series,experiments compare) or from the run's own output files. Never plot a remembered, rounded or plausible number, and never leave demo data in a script that ships. - The caption is the title. No axes title on a paper figure; panel letters (a, b) name the parts of a multi-panel figure.
- Show the uncertainty, or say there is none. One seed is an anecdote. Plot the interval across seeds and state the seed count in the caption; with one run, write "single seed".
- Label axes with units. "Loss" is a label; "step" without saying whether it counts optimizer steps or tokens is not.
- Colorblind-safe, greyscale-safe. Use the module's Okabe-Ito palette. Never
jet,rainboworhsv; they invent structure the data does not have. Baselines and chance levels are grey: color belongs to the things being compared. - One sans-serif face across every figure, diagrams included.
use_style()sets it. Passuse_style(family="serif")only for a figure carrying heavy math on a serif page.
What ships with it
6 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 22d ago Changed · +1 lines · -3 tokens per session 2d2b3294dd41
- 25d ago First seen · 141 lines · 102 tokens per session scan A 18e6dc78d70e
figures is a skill published in the GitHub repository synthetic-sciences/openscience (3,948 stars, last pushed yesterday), licensed Apache-2.0. It adds 99 tokens to every session and 1,868 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-14.
Other skills, from other repositories
paper-length-gate
Deterministic artifact-backed manuscript readiness gate for meta-paper-write. Validates the workspace LaTeX artifact before compilation while leaving final page-count enforcement to compilepdf.
paper-plot-stub
Plot a results CSV (x, ybaseline, yours) as a two-line matplotlib chart and write a PDF. Demo-only.
nano-pdf
Edit PDF text/typos/titles via nano-pdf CLI (NL prompts).
ocr-and-documents
Extract text from PDFs/scans (pymupdf, marker-pdf).
extracting-lab-tables
Detects and extracts tabular laboratory panels from PDFs, scans, and images into structured rows ready for OpenMed and FHIR. Use when the user has a CBC, CMP, lipid panel, or other lab report as a scanned image / PDF / spreadsheet and needs the test name, value, unit, reference range, and abnormal flag as clean rows.…
pdf-summarizer
Summarize the key points of a PDF document into a short bullet list.