{"uid":"cap_ZeAtSf0kdGvyznuVBXEc1","slug":"dag-markov-equivalence-skeleton-collider-witness-checker-ab60ecf4","name":"DAG Markov Equivalence & Skeleton/Collider Witness Checker","description":"A graphical-model test agent needs formal equivalence of two supplied DAG conditional-independence models, plus exact missing skeleton/collider witnesses. Not comparing graph isomorphism or a single d-separation query.","url":"https://detect.halowerk.com/v1/F-NIGHT-ECMA/dag-markov-equivalence-skeleton-collider-witness?utm_source=zero.xyz","method":"POST","headers":{},"bodySchema":{"type":"object","properties":{"artifact":{"type":"object","required":["markov_dags"],"properties":{"markov_dags":{"type":"object","required":["vertices","left","right","claim"],"properties":{"left":{"type":"array","items":{"type":"array","items":{"type":"integer","maximum":15,"minimum":0},"maxItems":2,"minItems":2},"maxItems":128,"minItems":0},"claim":{"type":"boolean"},"right":{"type":"array","items":{"type":"array","items":{"type":"integer","maximum":15,"minimum":0},"maxItems":2,"minItems":2},"maxItems":128,"minItems":0},"vertices":{"type":"integer","maximum":16,"minimum":1}},"additionalProperties":false}},"additionalProperties":false}}},"responseSchema":null,"example":null,"exampleRequest":null,"tags":["x402"],"displayCostAmount":"0.003","displayCostAsset":"USDC","priceDynamic":false,"priceHint":null,"priceStatus":"priced","priceSource":"probe","requiresHandshake":false,"reviewCount":0,"rating":{"score":"0.00","successRate":"0.00","reviews":0,"stars":null,"state":"unrated"},"availabilityStatus":"unknown","priceObserved":null,"sessionDeposit":null,"pricing":{"kind":"static","summary":"$0.003/call","primary":{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.003","per":"call","confidence":"exact"},"accepted":[{"kind":"static","protocol":"x402","network":"base","amountUsd":"0.003","per":"call","confidence":"exact"}]},"paymentMethods":[{"uid":"pm_knLebWSq7ye_XgeHoY-Qx","protocol":"x402","methodType":"crypto","chain":"base","mode":"charge","costAmount":"0.003","costPer":"request","priority":0,"asset":"0x833589fCD6eDb6E08f4c7C32D4f71b54bdA02913","unit":"request","depositMicros":null,"planRef":null}],"brandName":null,"brandSlug":null,"brandBaseUrl":null,"brandDocsUrl":null,"whatItDoes":"Formally tests whether two DAGs are Markov equivalent and identifies exact missing skeleton edges and collider witnesses that explain any difference.","exampleAgentPrompt":"Check whether these two DAGs (each up to 16 vertices) are Markov equivalent — left graph edges: [[0,1],[1,2],[0,2]], right graph edges: [[0,1],[2,1],[0,2]], vertices: 3, claim: true — and if they're not equivalent, give me the exact missing skeleton edges and collider witnesses that prove it.","exampleUseCases":[{"title":"Causal model equivalence audit","prompt":"I have two Bayesian network structures over 10 variables — left DAG edges [[0,1],[1,3],[2,3],[3,4]] and right DAG edges [[0,1],[1,3],[2,3],[4,3]], with 10 vertices total — I'm claiming they're Markov equivalent (claim: true). Can you formally verify that and show me any skeleton or collider witnesses if they're not?"},{"title":"Structural causal model debugging","prompt":"My pipeline generated two candidate causal graphs over 8 nodes: left edges [[0,2],[1,2],[2,3]] and right edges [[0,2],[1,2],[3,2]], and I think they're NOT equivalent (claim: false). Can you confirm and give me the precise collider witnesses that distinguish them?"},{"title":"Automated equivalence class testing","prompt":"For a research experiment I need to verify 16-vertex DAGs — left edges [[0,1],[1,2],[2,3],[3,4]] and right edges [[0,1],[1,2],[2,3],[4,3]] — are in the same Markov equivalence class (claim: true), and if not, list every missing skeleton edge and collider witness."}],"resultDescription":"Returns a verdict on whether the two supplied DAGs are Markov equivalent, along with exact witnesses: missing skeleton edges (undirected adjacencies present in one graph but not the other) and collider (v-structure/immorality) differences that constitute formal counterexamples to any incorrect equivalence claim.","failureModes":["Edge indices exceeding the declared vertex count — validation error","Vertex count outside 1–16 range — schema rejection","Edge arrays exceeding 128 edges — schema rejection","Malformed edge pairs (not exactly 2 integers) — schema rejection","Cycles in the DAG input (not a valid DAG) — may return error or undefined behavior"],"whenToPreferThis":"Use this endpoint when you need formal, exact Markov equivalence checking between two DAGs — not just a single d-separation query or graph isomorphism test. It is specifically designed for graphical-model agents that need both a yes/no equivalence verdict AND the precise missing skeleton edges and collider witnesses (immoralities) that explain any non-equivalence, making it suitable for causal discovery pipelines, structural equation model validation, and automated equivalence class enumeration.","instructions":null,"reviewSummary":null,"reviewSummaryHighlights":null,"reviewSummaryConcerns":null,"reviewSummaryGeneratedAt":null,"activationCount":0,"lastUsedAt":null,"lastSuccessfullyRanAt":null,"lastHealthCheckAt":"2026-11-20T00:37:29.332Z","isFirstParty":false,"canonicalSlug":"dag-markov-equivalence-skeleton-collider-witness-checker-ab60ecf4"}