Jak używać
- Skopiuj polecenie instalacji i sprawdź źródła przed instalacją we własnym środowisku agenta.
- W oryginale sprawdź wymagane narzędzia, dane dostępowe i obsługiwane środowiska.
- Opisz dane wejściowe i oczekiwany rezultat; porównaj wynik ze źródłami.
npx skills add https://github.com/yogsoth-ai/de-anthropocentric-research-engine --skill validate-causal-linkPrzed rozpoczęciem
- Instalacja nie obejmuje subskrypcji modelu, limitu API ani uprawnień do plików.
- Gwiazdki i instalacje pokazują aktywność, a nie niezależny test jakości lub bezpieczeństwa.
Oryginalny dokument Skill
Treść z repozytorium z zachowaniem nagłówków, przykładów, kodu, tabel, linków i obrazów.
validate-causal-link
Purpose
Audit a causal relation for clarity, existence, sufficiency, and logical completeness using CLR-style checks.
Input contract
yaml
required: [causal_relation, source_evidence, boundary_conditions]
optional: [causal_graph, alternative_explanations, clr_checklist]
constraints: [the relation has explicit cause and effect; distinguish clarity, existence, sufficiency, and logical completeness]Procedure
- Parse the causal relation and define its scope and terms.
- Check whether the stated cause exists and is connected to the effect.
- Test sufficiency, omitted conditions, and alternative explanations.
- Record each CLR result and return a validated, weakened, or unresolved link.
Output contract
yaml
produces: [causal_link_assessment, clr_results, missing_conditions, evidence_requests]
delta_fields: [findings, evidence_updates, hypothesis_updates, uncertainties]Quality gates
- All eight CLR categories are retained when the caller supplies the full checklist.
- Clarity, existence, sufficiency, and logical completeness are reported separately.
- A causal label is not accepted when evidence supports only association.
Parameterization
The caller must provide the cause/effect schema, evidence and boundary conditions, CLR checklist, alternative-explanation fields, and verdict vocabulary.
Failure and counterexamples
Reject links with undefined variables, missing scope, circular support, or causal language unsupported by the evidence.
Provenance map
- resolved: deep-insight/clr-validation
Preserved source criteria ledger
| source | physical line | kind | source criterion |
|---|---|---|---|
| deep-insight/clr-validation | 13 | structural | Retain eight CLR validation categories for causal claims: clarity, existence, sufficiency, and logical completeness are explicit checks. |
