There is a lot of 5e content to automate, so all contribution is welcomed and appreciated. Read the below carefully to get started.
- Fill out effects in Foundry (see Automation philosophy).
- Convert automated items from Foundry to here (see Development-guide).
- Review and test newly submitted items before an update (see Setting up).
- Learn how the module works and support users with troubleshooting (see Coordination).
- Simply use the module and let us know when there's a problem! 😊
We organise on the 5etools Discord server.
Although we will respond to issues and pull requests here, the easiest way to get a quick response or report minor typos/bugs is to message the #plutonium-general channel. Come join us if you'd like to help out! (Especially join us if you plan to fill out a lot of data at once—helps avoid wasted effort.)
Run these precursory steps when you first set up:
- Install Node.js.
- Clone this repo locally.
- Run
npm i
within your local repo.
To build the module based on the current data files, run npm run build
. The module will be outputted to dist/
, from which you can copy to your Foundry modules
directory (for Unix users, see transfer.sh
).
Use npm t
to verify the data files against the schema.
There isn't yet a tutorial to explain the data format. However, you can likely work most of it out by looking at examples. Familiarity with 5etools' homebrew format and Foundry's data structure (including for dnd5e, DAE, etc.) is very helpful.
The module/data/
directory is laid out as follows:
- Each 'entity type' (the array names in 5etools' JSON format—
"monster"
,"spell"
,"classFeature"
, etc.) is given its own directory. - Within that directory, a
__core.json
file contains data for all entities which are natively available on 5etools (i.e. without loading homebrew). - Within the same directory, each homebrew source has its own file named
<brewSourceJson>.json
(e.g.WJMAiS.json
for Wildjammer: More Adventures in Space). If a homebrew source has multiple datatypes, one file per datatype is required (excluding datatypes without automation).
Many facets of automation must be handled with macros. These are attached using DAE's "DIME", and appear as an escaped string in data. To add some human-readability, macros are handled separately and merged during the module build.
In the macro-item/
directory is a directory for each datatype. Save your (well-formatted, commented) macro code as a Javascript file in one of these directories, structured as an async function named macro
, with filename <sourceJson>_<item-name-lowercase-hyphenated>.js
(e.g. XGE_toll-the-dead.js
). Note that the first and last lines of the file—the ones that turn the macro into an async function—are stripped on compilation into the module's data.
You can create a new macro file using the npm run mt --
command. The directory (-d
) should match that of the JSON file into which the macro will be built.
npm run mt -- -d <directory> -s <jsonSource> -n <entity name>
# example: npm run mt -- -d spell -s PHB -n "toll the dead"
# example: npm run mt -- -d feat -s PHB -n "war caster"
To reference this human-readable macro in the main data files, use "itemMacro": "<filename>"
(e.g. "itemMacro": "XGE_toll-the-dead.js"
).
In tool/
(runnable locally with npm run local-pages
) is a simple webpage which you can use to automate some of the data-filling work. On the left, enter a Foundry item's JSON, or load a file at the top. The data will be stripped to just Plutonium Addon: Automation-compatible data (plus anything not recognised—sort through that yourself).
You can also upload entire items.db
files, but be aware that the webpage doesn't split the items out by (game-mechanical) datatype, and some modules might be hiding some clutter in that file (e.g. DFCE custom CEs).
In the interests of consistency and user experience, we have some guidelines to follow when adding content.
- We play D&D 5e.
- Always assume the default ruleset.
- If you can do some fancy (invisible) macro work to support variant rules, go ahead, but never compromise core function.
- Maximum efficiency; minimum effort.
- If something can be automated reliably, it should.
- If something can't be automated reliably, it shouldn't.
- Avoid automating only half of an item's effects without somehow informing the user of this.
- Avoid automating anything that isn't technically required, even if it feels obvious. For example, don't set the effects of the hold person spell to expire on a short rest, even though 1 hour is more than 1 minute. (If the user wants time-based AE-expiry, they can use another module like about-time.)
- A caveat to the above is when there is a false choice. For example, making the saving throw to end the hold person spell is technically optional, but since the target can literally do nothing except roll the saving throw (by virtue of paralysis), there is virtually no reason to not do so.
- Don't reinvent the wheel.
- Use the required modules wherever possible; don't use a macro unless you need to.
- Only introduce a new module if it's compatible, stable, and actively maintained. Finding replacements when a helper module goes kaput is a lot of effort.
- Use DFred's Convenient Effects for all conditions, activated via the
statusEffect CUSTOM <condition>
effect change. - Remember Midi QoL
overTime
effects exist.
- KISS: keep it simple, sweetie.
- Avoid using macros except when absolutely necessary, and make sure they're readable and maintainable.
- Certainly avoid referencing anything outside the item that you can't guarantee will be present.
- Plutonium should only import one item per game-mechanic name. If a single class feature or spell has multiple, independent functions (e.g. the paladin class' Lay on Hands feature), activating that item should prompt the user to choose the function (via a macro).