Useful, accurate, and safe at the same time
A card is publishable when it satisfies three tests, in this order. Accurate: every claim is verified against a primary source named in the card record. Safe: no wrong choice would be dangerous if a player misremembered it as the correct answer. Useful: the card teaches a real household decision, not abstract trivia.
Accuracy without safety is bad design. Safety without usefulness is timid design. We require all three.
Timeframe, hazard, domain, decision
Every card combines four elements. The timeframe sets when the card is relevant: the first 72 hours, the first two weeks, the recovery period, or long-term self-reliance. The hazard is one of 24 canonical disaster types. The domain is one of 15 preparedness pillars — water, food, energy, shelter, communications, medical readiness, transportation, planning, and seven more that unlock in later tracks. The decision is what the player is asked to choose between, drawn from a real situation a household actually faces.
A card that doesn't fit this formula isn't a card. It might be a Field Note, an info page, or a hub article. But it's not a Ready Quest card.
Primary sources only
Every safety claim must come from a primary source. The hierarchy: U.S. federal agencies (CDC, FEMA, NOAA, USGS, USDA, EPA, CPSC, FDA, ATSDR, NRC). Major standards bodies (American Red Cross, NFPA, AHA, NIOSH, OSHA). Peer-reviewed literature. State or provincial emergency management for jurisdiction-specific guidance.
We read the actual source page, note the specific number or procedure, and capture the URL with the card. If we can't find a primary source for a safety claim, we don't write the card. We park it for later research.
Every distractor gets a safety review
This is the most important section of the standard. For every wrong choice, we ask: If a player skimmed this choice, forgot which one was wrong, and remembered the wrong choice as the correct answer, would that be dangerous? If the answer is yes, the card is not publishable as written.
Wrong choices that fail by omission, overreaction, or out-of-date folk wisdom are safe to include — "Throw out everything that touched the water" is wasteful but not lethal. Wrong choices that name active dangerous behaviors are not — "Run a generator in the garage with the door cracked" is the kind of thing that kills people every winter, and we will not put that sentence in a list a reader might skim.
The wrong version. Question: "What is the safest way to heat your house during a winter outage?" Choices: kerosene heater in a small closed room · propane camp stove with a window cracked · generator in the garage with the door open · wood fireplace with a clean chimney and a CO alarm.
Three of the four wrong choices are lethal behaviors. A reader who skims a choice list cannot reliably remember which one was wrong. This card teaches one safe action while planting three dangerous mental images.
The corrected version. The question now names the equipment the household actually has and asks which is safest. The correct answer is qualified ("if the chimney is clear, with a battery-powered CO alarm running"). The teaching note then names each dangerous alternative explicitly, with the specific reason it kills people. The reader leaves with four specific facts, not three blurred ones.
When the lesson is the danger — carbon monoxide, driving through floodwater, generator placement — we mark the card as myth-fact type and put the danger in the question itself, not buried in the choice list. The reframe pushes the warning out of distractors and into framing.
Explain why, not just what
Each card ends with a 2-3 sentence teaching note, under 80 words. It does three things. It explains why the right answer is right, not just that it's right. It carries the source's calibration ("usually," "in most cases," "always") without inventing certainty. It names one edge case if there is one — altitude, age, jurisdiction.
The teaching note is where the learning happens. Without it, the card is a quiz; with it, the card is a training.
Five honest levels
Cards rate from 1 (Aware: universal knowledge) to 5 (Self-Reliant: judgment and wisdom). The editorial difficulty is set by the author. As community plays accumulate, an effective difficulty is computed from the success rate. When the two diverge by 2+ levels, the card goes back for editorial review — either the difficulty was misjudged, or the card has a problem.
A level-5 card that's just trivia-style memorization is miscategorized. Genuine level-5 cards turn on judgment under uncertainty.
What Ready Quest will not turn into trivia
- Personalized medical, financial, or legal advice. Cards stay at the level of universal household guidance. We don't tell anyone what their specific medication regime should look like during a disruption — we point them to the right professional.
- Jurisdiction-dependent answers without naming the jurisdiction. If "the right thing to do" depends on which state you're in, the question must name the state or the card doesn't ship.
- Brand-specific recommendations. We name generic categories (NOAA Weather Radio, hand-crank flashlight) but not specific products inside a card. Product reviews live on the New World Survival shelf, not in the game.
- Politicized framing. Disaster preparedness isn't a political identity. Cards do not lean toward any audience subculture.
- Prepper-opinion questions. "Is it better to bug out or shelter in place?" depends on the disaster. We don't traffic in the doctrinal debates of the preparedness subculture.
Two passes before publish, ongoing after
Every card requires a second-reviewer pass. The second person opens every source URL, confirms each cited fact against the primary document, and re-reads the distractors with the §6 safety review applied fresh. Reviewer initials populate the card's reviewer field.
Cards expire. The lastVerified date is set at publish; cards past 12 months on a safety-critical topic are flagged for re-verification before being served to players.
How we use AI tools, and how we don't
AI assists with drafting, structure, and second-pass review. Every safety-critical claim is then verified by a human against a primary source. AI may not be the only check on whether a wrong answer is dangerous to include — that's the §6 review, and it requires a person.
Cards in the current library are flagged with reviewer: "pending" until the human second-pass is complete.
Tell us, and we'll re-verify
If something looks off — out of date, contradicts a source you trust, or fails the §6 review — write to us. Reference the card id (visible at the bottom of every card during play) and what you noticed. We'll re-verify against the cited source and either correct the card or document the disagreement.
The point of the standard isn't that we never make mistakes. The point is that mistakes get fixed in public, with sourcing, instead of buried.