EcoKure / Our results
Our own results, connected to the records behind them. Inspect what each run measured and how far the evidence goes.
Results / Inspect the measurement
Choose one of EcoKure’s recorded runs. The picture describes that run; the linked source defines its scope.
Select an illustrated object to inspect its role.
01 / 09
1,000,000 records · 16,950 affected · 983,050 preserved.
Recorded internal / local synthetic runSynthetic in-process benchmark. It is not a production SLO, certification, customer result or comparison against Temporal/Camunda.
records avoided during replay
1,000,000 records · 16,950 affected · 983,050 preserved.
Synthetic in-process benchmark. It is not a production SLO, certification, customer result or comparison against Temporal/Camunda.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
synthetic iterations recorded
7/7 declared checks passed. Replayable affected work is recalculated; unknown provenance and human judgement are held for review.
Local AML/CTF threshold-change fixture with a declared dependency graph. This is not a customer deployment, legal interpretation or proof of completeness for an unknown graph.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
incorrect arithmetic claims blocked
20,000 synthetic cases · 2,987 abstentions · 0 incorrect claims missed.
The gate checks that a claimed result equals the expression it is claimed for. It does not check that the expression is the right expression, and an exactly-correct answer to the wrong question passes. Anything outside its four-operator grammar is abstained rather than passed, and that count is reported rather than folded into the refusals. The expression is parsed, never executed. Synthetic cases.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
invalid unit statements blocked
20,000 synthetic cases · 2,964 abstentions · 0 invalid statements allowed.
Dimensional consistency is necessary, not sufficient: a statement whose units agree can still be clinically wrong, and this gate says nothing about whether the number is right. It checks units only. Units outside its vocabulary are abstained rather than passed, and those abstentions are reported in two kinds: a vocabulary gap, which falls as the table is extended, and a notation that is not a dimensional quantity at all, which does not. Temperatures are treated as absolute; the gate does not distinguish a temperature from a temperature difference. Synthetic cases; no patient or clinical data.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
cases passed
17 safe refusals · 10 adversarial blocks · 0 failures. Run pbx_9700357081d0a2c9.
SuperMath ProofBench X tests deterministic mathematical verification, reproducible certificates, safe refusal and DTL warm-lane acceleration. It is not a model-only reasoning benchmark. Internal suite; no external ranking claimed.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
synthetic records classified
20 rounds · 504 affected · 683 preserved · 13 unknown.
Local synthetic changes to control, model and source references. No production gain or universal dependency-coverage claim.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
declared workflow checks passed
Onboarding, human-approved mapping, control release, selective replay, review and signed evidence were checked in a synthetic workflow.
Local synthetic full control loop; no external actuation. It does not establish tenant isolation, enterprise release approval or regulated-system qualification.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
tested mutations rejected
Payload, sequence, signatures, hashes, inserted/deleted records and key substitution were checked.
Local synthetic test. Integrity verification establishes that a record is unchanged; it does not establish scientific correctness or production security.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
repeat runs completed
2,000 records per round · 20 validation scenarios · 1 distinct validation digest.
Local synthetic repeatability evidence only; no customer data or production credentials. Wall-clock timings are diagnostic and are not a customer SLO. This does not establish production approval, regulatory compliance or absence of implementation defects.
Inspect the original record for the method and negative outcomes. Reproduce within its declared scope before evaluating another environment.
Research & release evidence
E-OSA / E-VSC / E-SCE and wider verified-reuse research have a different evidence boundary from these public gate and replay runs. No figure here establishes universal computational reuse.
The bounded E-ACC / ERA audit architecture is documented separately. Mandatory freshness and action-boundary claims require the evidence for that exact implementation and release; a website workflow test does not stand in for it.
Partner-reported results, independently reproduced results and production-qualified results require explicit provenance.
How to review
Check the case counts, outcomes, environment and reported scope.
Inspect the programme, open questions, corrections and reproducibility instructions.
A bounded pilot evaluates your inputs and baseline in the intended environment.
Continue your review
Browse all published claims, programme records and corrections.
