NEOCARIDINA READER COMPREHENSION MODERATOR KIT
Version: 1.0
Reviewed: 2026-08-12

SESSION ORDER
1. Record the scenario id, recruitment source category and declared experience band.
2. Read only the participant prompt. Do not reveal the rubric or preferred wording.
3. Record the initial response verbatim or as a faithful de-identified transcription.
4. Provide the declared guide paths without explaining their answer.
5. Ask the participant what they would do next, what information supports it and what remains unknown.
6. Score all three dimensions, prompting and the scenario-specific critical failure before debriefing.
7. Remove accidental private information and record the editorial decision.

FIXED RUBRIC
- Bounded next action: The response names an observable, measurable or reversible next step that fits the scenario without adding an unsupported intervention.
- Evidence or measurement: The response identifies the record, comparison, method or observation needed to support that action rather than substituting appearance, time or popularity.
- Nearest unsupported inference: The response states the closest tempting conclusion that the scenario and guide still cannot establish.

SCENARIO 1: commissioning-calendar
Audience: preparing
Participant prompt: Your planted tank has been running for six weeks. Ammonia and nitrite are currently reported as zero, but no known input was recorded. You hope to add shrimp this weekend. What would you do next, what information would you use and what can you still not conclude?
Guides: https://moltandmoss.com/care/cycling, https://moltandmoss.com/care/testing
Action boundary: Delay stocking and verify the whole operating system with a recorded fishless challenge using correctly identified total-ammonia and nitrite methods.
Evidence boundary: Use the known challenge input, test identities and reporting bases, timed ammonia and nitrite results and the system conditions rather than calendar age alone.
Limitation boundary: A completed challenge establishes only the processing demonstrated under its stated conditions; it does not certify every future load, grazing readiness or lifetime safety.
Critical failure code: calendar-as-proof
Critical failure: The response clears the tank from age, appearance, nitrate presence or one unchallenged zero reading.

SCENARIO 2: equal-tds-transfer
Audience: preparing
Participant prompt: The seller water and your aquarium both read 220 on meters labelled TDS. The shrimp have just arrived. What would you do next, what information would you use and what can you still not conclude?
Guides: https://moltandmoss.com/care/parameters, https://moltandmoss.com/care/acclimation
Action boundary: Keep the package protocol intact, compare actual available source and receiving measurements and choose a transfer plan that does not treat one matching display as matching water.
Evidence boundary: Use meter method and units, temperature, GH or other measured chemistry, package condition and chemistry where available, receiving-water records and the exact transfer trajectory.
Limitation boundary: Equal conductivity-derived TDS estimates do not identify dissolved substances, prove equal chemistry or establish one ideal acclimation duration.
Critical failure code: measurement-substitution
Critical failure: The response treats equal TDS displays as proof of equal composition or automatic transfer safety and does not seek any additional package or receiving-water context.

SCENARIO 3: new-source-separation
Audience: preparing
Participant prompt: A healthy-looking group from a new breeder can use a spare tank, but the only siphon and net are also used in your established colony. What would you do next, what information would you use and what can you still not conclude?
Guides: https://moltandmoss.com/care/quarantine, https://moltandmoss.com/care/acclimation
Action boundary: Maintain a separate source contact network with a commissioned system and dedicated wet equipment before receiving or releasing the new animals.
Evidence boundary: Use source identity and history, dedicated-equipment records, observations over time, transfer records and a stated release decision rather than appearance alone.
Limitation boundary: Separation reduces contact pathways but does not prove that animals are pathogen-free or make one universal quarantine clock valid.
Critical failure code: shared-equipment-called-quarantine
Critical failure: The response calls the setup quarantine while knowingly sharing wet equipment without a validated separation procedure.

SCENARIO 4: sudden-colony-change
Audience: first-colony
Participant prompt: Several shrimp are moving abnormally and two died after routine maintenance. A forum recommends dosing a broad treatment immediately. What would you do next, what information would you use and what can you still not conclude?
Guides: https://moltandmoss.com/care/health, https://moltandmoss.com/care/chemical-safety
Action boundary: Begin immediate system triage, support oxygenation when appropriate, preserve samples and build a time-ordered record before adding an unidentified treatment exposure.
Evidence boundary: Use temperature, oxygen context, correctly labelled nitrogen results, affected and comparison denominators, recent water and product inputs, timing, anatomy and behavior.
Limitation boundary: The pattern and one community recommendation do not diagnose an agent, identify a chemical cause or authorize a treatment or dose.
Critical failure code: treatment-before-triage
Critical failure: The response recommends or begins a treatment or dose before basic system triage and identification boundaries are addressed.

SCENARIO 5: peaceful-fish
Audience: considering
Participant prompt: A store calls a small fish peaceful and says adults usually coexist with shrimp. Your goal is a colony that produces juveniles. What would you do next, what information would you use and what can you still not conclude?
Guides: https://moltandmoss.com/care/tankmates, https://moltandmoss.com/care/stocking
Action boundary: Separate adult display from recruitment goals and choose a dedicated shrimp system unless evidence for the exact fish, size, density and habitat supports the desired outcome.
Evidence boundary: Use exact fish identity and size, shrimp life stages, habitat and cover, complete adult and juvenile counts and repeated-brood outcomes rather than a peaceful label.
Limitation boundary: Visible adult coexistence does not establish hatchling retention, juvenile recruitment, welfare or universal compatibility.
Critical failure code: adult-coexistence-as-recruitment-proof
Critical failure: The response treats adult survival or a peaceful trade label as proof that juveniles will recruit and does not separate adult display from a repeated-brood colony outcome.

SCENARIO 6: leftover-food
Audience: first-colony
Participant prompt: Food remains visible several hours after feeding, and someone tells you every shrimp colony should receive the same pellet count each day. What would you do next, what information would you use and what can you still not conclude?
Guides: https://moltandmoss.com/care/feeding, https://moltandmoss.com/care/nitrogen
Action boundary: Remove or manage leftovers according to the recorded method, reduce the next measured ration and observe consumption and system response before another adjustment.
Evidence boundary: Use exact food and lot when known, dry or wet portion, animal and life-stage context, delivery location, timed consumption, leftovers, background grazing and water observations.
Limitation boundary: One feeding response does not establish nutritional completeness, an ideal ration, a universal pellet count or the cause of a later water result.
Critical failure code: universal-feeding-clock
Critical failure: The response preserves or replaces the ration with another universal pellet count or calendar without using observed consumption.

SCENARIO 7: white-ring-record
Audience: first-colony
Participant prompt: A photograph shows a pale ring between body sections before one shrimp dies. A comment says this proves low calcium. What would you do next, what information would you use and what can you still not conclude?
Guides: https://moltandmoss.com/care/molting, https://moltandmoss.com/care/health
Action boundary: Preserve the photograph as one observation, record the event and check the broader system and molt history before changing minerals or assigning a cause.
Evidence boundary: Use timing, anatomy, movement, recent molts, affected denominator, comparison animals, measured water with methods and units, food and recent inputs.
Limitation boundary: A visible pale ring does not diagnose calcium deficiency, prove a failed molt mechanism or show that raising GH is the correct intervention.
Critical failure code: visible-sign-as-diagnosis
Critical failure: The response diagnoses mineral deficiency or recommends changing hardness from the photograph alone.

SCENARIO 8: repeated-community-pattern
Audience: established
Participant prompt: Ten keepers report that the same leaf made their colonies more active, but their tanks, portions and observation times differ. What would you do next, what information would you use and what can you still not conclude?
Guides: https://moltandmoss.com/care/habitat, https://moltandmoss.com/care/behaviour
Action boundary: Preserve the reports as a question-refining signal and define one identified preparation, endpoint and prospective observation method before seeking comparable records.
Evidence boundary: Use plant identity and source, preparation, portion, background feed, tank context, observation time, denominator, contrary cases, missing data and the exact claimed endpoint.
Limitation boundary: Repeated unassigned reports do not establish ingestion, nutritional value, health benefit, causation, safety or a universal botanical recommendation.
Critical failure code: popularity-as-biological-proof
Critical failure: The response treats repetition or agreement as causal proof or recommends the leaf broadly without defining preparation and endpoint.

BLANK SESSION RECORD
Session date:
Recruitment source category:
Experience band:
Scenario id:
Initial response:
Post-guide response:
Action dimension present: yes / no
Evidence or measurement dimension present: yes / no
Nearest limit dimension present: yes / no
Moderator prompting used: yes / no
Critical failure present: yes / no
Failure codes:
Outcome:
Accidental private information removed: yes / no
Editorial decision:
Reviewer: