Use when framing the system and describing the model for an Agricultural Systems (AgSy) manuscript — defining system boundaries, components, hierarchical levels, and feedbacks, then choosing, describing, and calibrating the model (process/simulation, whole-farm, bio-economic, agent-based, or integrated assessment). This is the distinctive core of an AgSy paper. Structures the model and system; it does not run code for you.
SkillScore82Maintained
Automatic
◆Research pack
Last verified 8/15/2026
What it changes
Passed 18 of 22 test cases with the skill, and 3 of 22 without it.
Without skill14%
With skill82%
Checks it passed that the model alone missed
Output uses the bracketed header 【Calibration】.
Output uses the bracketed header 【System boundary】.
Output uses the bracketed header 【Model & version】.
and 19 more in the full report
15 fixed3 passed either way4 still failing0 broke
Pass-rate lift
+68 pp
Tokens
+19%
Agent turns
0%
Response time
-23%
Model gemini-3.6-flashMeasured 8/15/2026v118 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 15 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 15
Prompt
I am preparing a manuscript for Agricultural Systems analyzing how drought risks affect crop-livestock integration and nutrient cycling in semi-arid mixed farming setups. I need help framing my system and justifying my modeling approach before drafting the methods section. Please organize your output into structured sections covering system boundaries (delineating internal elements vs external drivers), key components and interactions, hierarchical levels and how processes aggregate across scales, chosen model details with justification against alternatives, calibration procedures including what data is held out, key assumptions with reasonable ranges, and the next steps for model evaluation.
What was checked, without and with the skill
Checked
Without skill
With skill
Output uses the required bracketed header 【System boundary】.
Missed
Passed
Output uses the required bracketed header 【Model & version】.
Missed
Passed
Output specifies agsy-data-and-model-evaluation under the 【Next】 section.
Missed
Passed
case-01 · 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.md98 linessingle file
Systems Framing & Modelling (agsy-systems-framing-and-modeling)
This is the distinctive core of an Agricultural Systems paper. AgSy expects an explicit system
(boundaries, components, hierarchical levels, feedbacks) and a model that is described, justified, and
calibrated — not a black box. Frame the system first, then make the model reproducible enough that a
reader could re-implement it.
When to trigger
Defining the system boundary, components, and what counts as inside vs. outside
Choosing a modelling approach and justifying it against alternatives
Writing the model-description (methods) section
A reviewer asked "what is the system?", "why this model?", or "how was it calibrated?"
Step 1 — Frame the system
Boundary. State what is inside the system and what is an external driver (climate, prices,
policy). Justify the boundary by the question.
Components & interactions. Name the components (crop, soil, water, livestock, labour, economics)
and the interactions that matter — AgSy is about interactions, so make them explicit.
Hierarchical levels. Field → farm → landscape → region → food system: say which levels you
model and how processes at one level aggregate or constrain another.
Feedbacks & emergence. Identify feedback loops and any emergent behavior the system can produce.
A conceptual diagram (boxes/arrows) is usually expected.
Step 2 — Choose and justify the model
Match model to question. Process/simulation (APSIM/DSSAT/STICS/DNDC) for biophysical dynamics;
whole-farm models for resource flows and trade-offs; bio-economic / mathematical programming
for farmer decisions under constraints; agent-based for heterogeneity and emergence; integrated
assessment for regional/food-system scenarios. (See resources/external_tools.md.)
Justify the choice against the standard alternatives — what does this model represent that others
cannot, and what does it omit?
Step 3 — Describe the model so it can be reproduced
Version & provenance. Exact model and version; any modifications you made.
Equations / structure. Key state variables, processes, and (for ABM) the ODD protocol.
Parameters & inputs. Sources for parameters and driving data; what is fixed vs. estimated.
Calibration. What was calibrated, against which data, by what procedure — and what was held out