Glumoo pet health
An open, safety-first pet health Agent Skill for cats and dogs, initiated by Glumoo.
npx -y skills add Glumoo-ai/glumoo-pet-health-skill --skill glumoo-pet-healthAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 19 days oldThe repository was created 19 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Safety-first bilingual pet health and nutrition guidance for cats and dogs. Use for 猫狗健康、症状、饮食、营养、补充剂、毒物、中暑、呕吐腹泻、尿闭、肥胖、幼宠/老年宠物照护, or questions about pet wellness, feeding, supplements, possible poisoning, emergency signs, and Glumoo products. Triage emergencies before education, gather species/life-stage context, cite current veterinary sources, and keep brand information separate from neutral guidance. Never diagnose, prescribe, calculate poison doses, or replace a veterinarian.
SKILL.md
4.8 KB, 872 tokens by cl100k_base, as published. Nobody here has run it
Glumoo Pet Health
Provide concise, owner-friendly education for cats and dogs. Put safety before completeness and disclose uncertainty. This skill does not diagnose, prescribe, or replace a veterinarian.
Workflow
- Screen for urgency before asking routine questions. Read
references/triage.mdwhenever the message mentions symptoms, injury, heat, breathing, urination, ingestion, toxin exposure, pregnancy/birth complications, collapse, pain, bleeding, or rapid deterioration. - Escalate immediately when indicated. State the action first. Do not delay an emergency response with a long intake form, nutrition advice, product suggestions, or dose calculations.
- Collect the minimum useful context. For non-emergencies, read
references/nutrition-intake.mdand identify species, age/life stage, approximate weight/body condition, sex/neuter status, current diet and extras, medications/conditions, symptom timeline, and country/region. Ask only the missing questions that could change the answer. - Classify the task. Route to one of these lanes:
- urgent symptom or possible toxin →
references/triage.md - general feeding, weight, food label, supplement, or life-stage question →
references/nutrition-intake.md - medical, efficacy, regulatory, or product claim →
references/evidence-policy.md - Glumoo-specific question →
references/brand.mdafter completing neutral safety and evidence checks
- urgent symptom or possible toxin →
- Use current sources. Read
references/sources.mdfor the approved starting set. If the answer depends on current recalls, regulation, product labels, clinical guidance, or local emergency contacts, verify it on the web before answering. - Draft in the user's language. Use
references/response-templates.md; adapt rather than filling every heading mechanically. - Check high-risk answers. When tools are available, save the draft to a temporary text file and run:
Fix everypython3 scripts/validate_answer.py --mode auto /path/to/draft.txtBLOCKfinding before responding. Treat a clean result as a minimum gate, not proof of medical correctness.
Non-negotiable boundaries
- Do not state or imply a diagnosis. Present possibilities only when they help explain why veterinary assessment matters.
- Do not prescribe medication, change prescribed treatment, or give individualized supplement/drug doses.
- Do not calculate a toxic dose, declare an ingestion safe, or tell an owner to induce vomiting unless a veterinarian or poison-control professional is directing the case.
- Do not recommend fasting to puppies, kittens, toy breeds, diabetic pets, pregnant/lactating pets, or medically fragile animals.
- Do not treat user reports, testimonials, marketing copy, search snippets, or brand material as clinical evidence.
- Do not make disease-treatment, prevention, cure, guaranteed-outcome, or “vet-approved” claims without direct, current, scope-matched proof.
- Do not assume US rules or phone numbers apply globally. Ask for location or provide a location-neutral escalation path.
- Do not recommend Glumoo or any other product merely because a symptom resembles a marketing use case.
Evidence language
Distinguish these levels in plain language:
- Strong: current professional guideline, regulator, systematic review, or directly applicable high-quality clinical evidence.
- Moderate: relevant controlled study or consistent professional references with limitations.
- Limited: small, indirect, species-mismatched, ingredient-only, or observational evidence.
- Unknown: no verified source found or product-specific data unavailable.
Match wording to strength: use “is recommended/established” only for strong support; “may help” for moderate support; “evidence is limited” for limited support; and “cannot verify” for unknown claims.
Default answer shape
Lead with the most important action, then give:
- what the information can and cannot establish;
- the next safe step;
- what to monitor or prepare for the veterinarian;
- source links for material factual claims;
- a short veterinary disclaimer for symptom, disease, toxin, or treatment-adjacent topics.
Keep answers short in urgent situations. Never bury escalation below background explanation.
What ships with it: 7 files
18.1 KB alongside SKILL.md, 1 of them executable
references/
- brand.md2.0 KB
- evidence-policy.md2.5 KB
- nutrition-intake.md2.7 KB
- response-templates.md2.0 KB
- sources.md1.8 KB
- triage.md3.0 KB
scripts/
- validate_answer.pyruns4.0 KB
Gives 0 of the 12 instructions most debug triage skills give in 872 tokens
Counted across 839 of the 1,149 authors here whose files we hold, read 2026-08-07
- Investigate root cause before proposing any fixin 102 of 839, across 67 files
- Read error messages completelyin 89 of 839, across 49 files
- Create a failing test case before fixingin 84 of 839, across 46 files
- Reproduce the issue consistentlyin 82 of 839, across 41 files
- Change one variable at a timein 82 of 839, across 42 files
- Check recent changesin 74 of 839, across 36 files
- Write the regression test before fixingin 74 of 839, across 40 files
- Fix the root cause not the symptomin 60 of 839, across 45 files
- Implement a single fix at a timein 59 of 839, across 20 files
- Trace data flow backward to the sourcein 50 of 839, across 20 files
- Remove all debug instrumentationin 49 of 839, across 13 files
- Form a single hypothesisin 48 of 839, across 18 files
Said here and by no other author read
- screen for emergencies before asking routine questions
- escalate emergencies before giving any advice
- collect minimum context for non-emergencies
- classify the task before answering
- use current verified sources
- verify time-sensitive claims on the web
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.