bioepic-data/ecosim-agent / .agents/skills/ecosim-trait-deriver
ecosim-trait-deriver
VerifiedUse this skill when deriving EcoSIM plant trait parameter recommendations for a named plant from web and online literature evidence, using a read-only `plant_trait.*.desc` file and the `ndlf43` tree block or `gr3s43` grass block as the starting archetype. Never edit or replace the `.desc` file.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.8 KB |
| references/web_trait_derivation.md | markdown | 4.1 KB |
| scripts/extract_trait_profiles.py | script | 5.8 KB |
references/web_trait_derivation.md
View on GitHub# Web Trait Derivation
Use this reference when the task is to derive EcoSIM parameters for a named plant from web and online literature evidence.
## Decision sequence
1. Resolve the accepted scientific name.
2. Decide whether the plant maps best to the tree anchor `ndlf43` or the grass anchor `gr3s43`.
3. Extract the anchor traits from the template `.desc`.
4. Search for plant-specific evidence in both species-profile sources and online literature.
5. Update the anchor values only where evidence supports a change.
## Recommended source stack
### Identity and growth form
- Kew POWO
- USDA PLANTS or USDA Forest Service
Use these to determine:
- accepted scientific name
- family
- life form
- perennial vs annual
- woody vs herbaceous
- broad climate or habitat context
### Trees and forestry species
- USDA Forest Service Silvics pages
- USDA Fire Effects Information System
- peer-reviewed forestry ecology papers
- online theses, dissertations, technical reports, and forestry books when they contain primary measurements
Useful for:
- height and growth form
- drought tolerance
- shade tolerance
- regeneration strategy
- root habit
- phenology
- habitat and climate range
### Grasses and crops
- FAO ECOCROP
- USDA crop resources
- peer-reviewed crop physiology papers
- online agronomy theses, extension technical bulletins, and cultivar reports when they contain measured traits
Useful for:
- life span
- habit
- photoperiod
- growing cycle
- climate requirements
- crop physiology and nutrient traits
## Online literature search workflow
For any quantitative trait that is likely to be measured in the literature, do not rely only on species summaries.
Search iteratively:
1. accepted scientific name + exact trait term
2. accepted scientific name + synonyms or older taxonomy + exact trait term
3. accepted scientific name + organ or process terms such as `photosynthesis`, `leaf nitrogen`, `specific leaf area`, `seed mass`, `root depth`, `hydraulic`, `water potential`
4. genus name + exact trait term if species-level search fails
5. functional-type synthesis only if species and genus searches fail
Good literature targets include:
- journal articles
- supplemental tables and appendices
- books and species monographs
- theses and dissertations
- technical reports with primary measurements
- trait datasets with traceable citations
When a value comes from a secondary source, prefer tracing it back to the primary citation if possible.
## Trait mapping heuristics
### Safe direct mappings
These can often be mapped directly from source descriptions:
- `ISTYP`
- `IDTYP`
- `IWTYP`
- `IPTYP`
- `MY`
- `ZTYPI`
- `WDLF`
- `GRMX`
- `GRDM`
- `WTSTDI`
### Usually inference-heavy
These should be changed only with stronger support:
- `VCMX`
- `VOMX`
- `ETMX`
- `RUBP`
- `UPMXZH`
- `UPMXZO`
- `UPMXPO`
- `RCS`
- `RSMX`
Traits in this group should usually trigger a literature search before you decide to keep the anchor.
### Often template-retained
Unless high-quality quantitative data are found, these often remain close to the anchor:
- `OPTICAL PROPERTIES`
- many `ROOT UPTAKE PARAMETERS`
- many `ORGAN GROWTH YIELDS`
Even for these, still check online literature if the user is asking for a species-specific derivation rather than a fast approximation.
## Output expectations
For each changed trait, record:
- trait code
- derived value
- rationale
- evidence source
- provenance label
Also record:
- whether the support came from species page, database, or online literature
- whether the value is species-level, genus-level, or functional-type-level
For unchanged traits, prefer:
- keep the anchor value
- mark it `template-retained`
## Examples
### Limber Pine
- likely anchor: `ndlf43`
- evidence to look for:
- evergreen conifer
- long-lived, slow-growing tree
- drought and cold tolerance
- deep roots / woody roots
- ectomycorrhizal association
### Maize
- likely anchor: `gr3s43`
- evidence to look for:
- annual C4 grass
- crop cycle and photoperiod response
- high nutrient demand
- shallow to intermediate fibrous root system
- crop physiology measurements from agronomy literature
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |