Use when building figures and tables for an Agricultural Systems (AgSy) manuscript so exhibits communicate interactions, dynamics, trade-offs, and model performance clearly. AgSy is a systems journal, so the best exhibits show trade-off frontiers, observed-vs-simulated fit, resource flows, and conceptual system diagrams — not just a bar chart of one treatment. Designs exhibits; it does not run the analysis.
SkillScore91Maintained
Automatic
◆Writing & Editing pack
Last verified 8/15/2026
What it changes
Passed 21 of 23 test cases with the skill, and 18 of 23 without it.
Without skill78%
With skill91%
Checks it passed that the model alone missed
The response mandates vector format export such as PDF or EPS.
The response specifies deploying the system boundary diagram as a graphical abstract or structural boundary exhibit.
The response recommends placing goodness-of-fit metrics such as RMSE or NSE directly onto an observed-versus-simulated plot.
and 1 more in the full report
4 fixed17 passed either way1 still failing1 broke
Pass-rate lift
+13 pp
Tokens
+41%
Agent turns
0%
Response time
-19%
Model gemini-3.6-flashMeasured 8/15/2026v121 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 4
Prompt
We need an exhibit for our submission to establish overall system scope, highlighting component boundaries and resource flows between livestock sheds, manure lagoons, and crop fields. What exhibit type is recommended, and where is it typically deployed in the manuscript?
What was checked, without and with the skill
Checked
Without skill
With skill
The response recommends a conceptual system boundary diagram with boxes and arrows.
Passed
Passed
The response specifies deploying the system boundary diagram as a graphical abstract or structural boundary exhibit.
Missed
Passed
case-07 · gemini-3.6-flash · 8/15/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md93 linessingle file
Figures & Tables (agsy-figures-and-tables)
Exhibits are where a systems reviewer checks whether the model and the trade-offs are real. At AgSy the
strongest exhibits make interactions, dynamics, and trade-offs visible, and let a reader judge
model performance at a glance. A wall of single-treatment bar charts does not belong in a systems
paper.
When to trigger
Designing the main results figure or a key model-performance exhibit
Deciding what belongs in the article vs. supplementary material
A reviewer found an exhibit unclear, mislabeled, or not self-contained
Presenting scenario comparisons or trade-offs
Principles
Self-contained. A reader should understand each exhibit from its caption, axis/column labels, and
note alone. State units, system level (field/farm/landscape/region), scenario, N, and time horizon.
Show the system, not one factor. Prefer exhibits that reveal interactions and trade-offs:
Trade-off / Pareto-front plots (e.g., profit vs. N leaching across options)
Observed-vs-simulated (1:1) plots with fit statistics (RMSE/NSE/bias) for model evaluation
Time series of state variables to show dynamics and feedbacks
Sankey / flow diagrams for resource, nutrient, or energy flows
Conceptual system diagrams (boxes/arrows) for boundaries and components
Multi-indicator / radar profiles for sustainability comparisons
Show uncertainty. Intervals, ensemble ranges, or boxplots — not bare point estimates from a
single model run.
Accessible. Colorblind-safe palettes; legible in grayscale; no chartjunk or 3D; vector output
(PDF/EPS) for print.
Reproducible. Each exhibit is generated by the master workflow; numbers match the deposited
data/code/model exactly (see agsy-reproducibility-and-data-policy).
Article vs. supplementary
Keep the few exhibits that carry the systems argument (trade-offs, key dynamics, model fit) in the
article; move full calibration tables, parameter lists, and exhaustive scenario grids to
supplementary material.
Pick the exhibit from the claim (illustrative)
A crop–livestock simulation has four candidate main figures; choose by what the systems claim needs:
Use when preparing the data, code, and model materials for an Agricultural Systems (AgSy) manuscript. AgSy applies Elsevier's research-data policy, which treats software, code, and models as research data — deposit them in a repository and cite/link them, or state why they cannot be shared. Covers model-description standards and exemptions. Prepares the materials; it does not waive requirements.