Write decisions and decision pools
Write the architectural decisions (ADRs) your dimension’s campaign presents players with, group them into pools, and optionally register your own reusable reward or penalty definitions — all in the CAMPAIGN tab’s DECISIONS, POOLS and EFFECTS sub-tabs.
Write a decision
Section titled “Write a decision”- In the DECISIONS sub-tab, click + ADD DECISION (or UPLOAD JSON for several at once).
- Pick Domain, Subtopic and Scope (act or mission).
- Fill in Title, Context (the realistic setup — no hint at the consequence) and Prompt.
- Optionally check Verified, and fill Tags, Author and Sources.
- Add 2 to 4 Options. Each needs a
key, alabel, a Quality (best, acceptable, poor), a Reversal class (two-way, refactor, migration, one-way), arationale, and an outcome: pick a REWARD and/or one or more PENALTIES, or write an inlineeffectsblock — never both. At least one option must be gradedbest. - Click CREATE.
The REWARD dropdown and PENALTIES checkboxes list every active platform effect plus your dimension’s own (see Register a reusable effect below). Picking a reward or penalty on an option uses that instead of its inline effects; bulk JSON upload supports the same either/or shape via the reward/penalties fields.
The outcome-band rubric
Section titled “The outcome-band rubric”Every option’s effect is checked against its quality’s band — write outside the band and the save is rejected:
| Quality | Positives | Structural negatives | Scheduled incidents | Slot ops |
|---|---|---|---|---|
best |
exactly 1 | 0 | 0 | up to 1 |
acceptable |
0 | 0 | up to 1, no earlier than turn 4 | 0 |
poor |
0 | exactly 1 | up to 1, no earlier than turn 4 | up to 2 (up to 2 incident-gate hexes if that’s the negative) |
A best option’s one positive is a buff, an opened shortcut slot, or exactly +2 Brain Power — never a softer negative alongside it. best and acceptable options also can’t carry an immediate soft negative (a debuff, an AI buff, or a Brain Power or turn-cap penalty); acceptable’s only permitted negative is its single deferred incident. Any option with an immediate visible effect needs a banner sentence. An incident gate’s subtopic must match the decision’s own subtopic.
Reversal class
Section titled “Reversal class”The door test: a config-level change is two-way; a code-structure change is refactor; a data or infrastructure change is migration; an external-contract change is one-way. Reversing a wrong decision costs demonstrated understanding, not currency — a Fix point of interest demands correct answers at a tier the class sets:
| Class | Questions to fix | Tier |
|---|---|---|
two-way |
1 | 3 |
refactor |
2 | 3 |
migration |
3 | 3 |
one-way |
4 | 3 |
Write a decision pool
Section titled “Write a decision pool”A pool is what an act or a mission’s ADR block actually draws decisions from at runtime.
- In the POOLS sub-tab, click + ADD POOL (or UPLOAD JSON).
- Fill in Name, Description, Domain and Scope (act, mission, mixed).
- Check the Subtopics this pool draws from.
- Set the Balance minimums — how many
best,acceptableandpooroptions, and how many net-positive and net-negative options, the pool as a whole must contain. - Optionally pin specific decisions so they’re always included.
- Fill the criteria JSON block (subtopics, tags, scope) — a decision matching it is pulled into the pool automatically, on top of anything pinned.
- Click CREATE.
When you save a pool, its resolved membership (pinned decisions, plus every decision matching its criteria) must be at least 4× the sum of its Balance minimums (at least 4 decisions), and must clear every balance minimum. When an act or mission references the pool, its membership must also be at least 4× that reference’s draw size. On top of your declared minimums, a pool can never resolve to all-reward or all-punishment — it always needs at least one of each.
Register a reusable effect
Section titled “Register a reusable effect”A named effect lets a decision’s option reference a reward or penalty by key instead of writing out its effects bundle inline, via the REWARD/PENALTIES fields on the decision form (or bulk JSON). Your dimension’s decisions can reference its own registered effects as well as platform ones.
- In the EFFECTS sub-tab, click + ADD EFFECT (or UPLOAD JSON).
- Set a unique Key (this is the effect’s identity — it can’t be changed after creation), Kind (
rewardorpenalty), Label, Description, Icon and Banner. - Pick an Effect kind — one of a closed set: RESILIENCE, THROUGHPUT, VELOCITY, CLARITY, BP, HARDENED_PATH, BLOCKED_PATH, GARRISON, INCIDENT_GATE. Your
Kindmust match this effect kind’s own polarity (arewardkind can’t point at a penalty-only effect kind, and vice versa). - Click CREATE.
Deleting, archiving or changing the kind of an effect still referenced by a decision is refused; reactivating an archived decision re-checks it against the current effect catalog.
Who can author
Section titled “Who can author”The dimension’s owner and admins. For a dimension owned by an organisation, that organisation’s owner and admins too.
Related
Section titled “Related”- Build missions and acts — where a pool gets drawn from.
- Bulk-upload content — the JSON shape for an option’s
reward/penaltiesfields, same either/or rule as the decision form. - How content is structured