| name | emiproc-developer |
| description | Use when developing emiproc internals: adding or changing inventories, exporters, regridding logic, profiles, scripts, and tests with maintainer-level quality checks for regressions, units, and API behavior. |
| argument-hint | task=<feature|bugfix|refactor> area=<inventories|exports|regrid|profiles> |
| user-invocable | true |
Emiproc Developer Skill
Mandatory First Step
Goal
Support maintainers and contributors making code changes in emiproc with minimal regressions and clear validation.
Use This Skill When
- Adding a new inventory reader or modifying an existing one.
- Extending export logic (NetCDF, rasters, model inputs).
- Implementing or improving a processing function (regridding, speciation, scaling, ...).
- Fixing bugs in emiproc.
- Preparing a PR with tests and docs updates.
Change Guidance By Area
In general make sure to read the developer tutorials in the docs.
Inventories
- Keep parsing and normalization logic explicit.
- Document assumptions about category and substance naming.
- Make sure to follow emiproc's internal data model conventions (for example units, category/substance organization in
gdf).
- If external data is freely available, add a function to download it.
Exports
- Preserve stable metadata and coordinate conventions.
- Verify required output units and variable naming consistency.
- Warn the user about potential limitations or incompatibilities in the export (example: if the model doesn't support temporal profiles as defined in the emiproc inventory).
Project Pointers
