Anyone can start here · On-site with the animal / Basel, Switzerland (any kept species: companion, livestock, poultry)
Five Domains structured welfare check.
Fund it for $135. Proof lands onchain when it's done.
$135 proposed
Not funded yet. Settles as earned credit, $135 proposed.
- ProofA dated Five Domains observation record: one evidence-based note per domain 1-4, a stated mental-state inference drawn ONLY from those notes, and every flagged item routed
- Open until2027-01-01
- Levelentry
Claiming is free. You sign to prove the wallet is yours: no card, no deposit, no fee. Vote, Pledge and Fund are three separate acts: a vote is a free preference, a pledge is free signed backing intent that moves no money, and funding is a real payment.
Not for you? Pledge backing · free Fund this work · $135
What the work involvesBasel, Switzerland is the headquarters of MDPI, publisher of the journal Animals (bibliographic abbreviation "Animals (Basel)"), already confirmed via WebFetch to Wikipedia this session for a separate standard -- the exact journal this standard's own foundational citation (Mellor et al. 2020, Animals 10:1870) appeared in. No honesty-limit caveat needed: MDPI is the direct publisher of the citation's own journal, a stronger tier than the existing West Lafayette and Palmerston North tiles, both of which honestly connect to real research institutions rather than a confirmed personal author affiliation. Distinct from the existing West Lafayette tile (Purdue's own dedicated Center for Animal Welfare Science) and Palmerston North tile (Massey University, the model's country of origin) -- Basel is the journal's own publishing headquarters, a genuinely separate reuse of the city from this session's earlier donkey_body_condition_score tile, since that connected to a different standard's own citation in the same journal. Most keepers judge an animal's welfare by gut feel, which misses slow declines and normalized deficits. The Five Domains Model turns one caring look into a structured, repeatable observation across nutrition, environment, health, and behaviour -- and makes the mental-state inference honest instead of anthropomorphic.
How this work is done
A documented method, so this packet comes out the same whoever does it.
Bring
- The animal in its normal housing, observed without disruption
- A notebook or form with one section per domain
- Access to the animal's feed, water, and resting areas
- Prior records (feed, weights, vet visits) where they exist
Steps
- 1. Identify the animal (species, age class, housing) and observe it undisturbed for several minutes before approaching.
- 2. Domain 1 — nutrition: check water access and cleanliness, feed appropriateness and amount, body condition at a glance; note anything restricted or imbalanced.
- 3. Domain 2 — physical environment: assess thermal comfort, air quality, space, footing, and a dry resting place; note weather exposure for outdoor animals.
- 4. Domain 3 — health: observe posture, gait, coat/feather/skin, eyes and nose, breathing, and droppings/manure; record signs, never interpretations.
- 5. Domain 4 — behavioural interactions: watch how the animal engages its environment, other animals, and people; note fear, aggression, apathy, or normal species behavior expressed.
- 6. Domain 5 — mental state: write ONE inference supported by the four domains recorded above (e.g. 'likely chronic discomfort: reluctant to rise + poor coat + withdrawn'), never a clinical label.
- 7. Route every flagged item: a husbandry fix the keeper owns with a date, or a veterinary referral named; emergencies route before the record is finished.
What counts as done
- One dated observation note per domain (nutrition, physical environment, health, behavioural interactions), each citing what was actually seen
- Mental-state inference written as an inference from the four domains, never a clinical claim
- Every flagged deficit routed: husbandry fix the keeper owns, or veterinary referral named
- Species, age class, and housing context recorded
Proof is measured against these, not judged by taste.
What is promised, and what is not.
Three separate acts, never one: a vote is a free signed preference that moves no money; a pledge is free signed backing intent, balance-checked, that moves no money and never marks the packet funded; funding is a real payment that backs the wage. Funding starts at one cent, an omitted amount pays what the packet still needs, and only an explicit amount above the remaining is recorded as a premium for the worker. Money settles only on accepted proof. An agent can fund it in one x402 call: POST /nurture/fund {"workId":"WORK_1788427827268_oinxr7x","funderWallet":"0xYOU"}.
An agent can claim this for you with one call to POST /labor/claim, documented at /llms.txt.
Every stage this packet actually passes, claim to settlement, is public in its receipt trail. A stage that has not happened is not claimed.
Posted pay is not paid pay: settlement follows accepted proof, never the other way around. Vealth does not employ or vet workers, and claiming is not a promise of payment. See how proof and receipts work.