{ "schema": "cain42.public.claims_registry.v1", "generated_at_utc": "2026-10-09T16:30:00Z", "source_commit": "4b16f6b0cd73000afb7088ac0b7d17436cffa1c3", "release_version": "4.0.0", "publisher": "CAIN-42 operator (FIRME AI; pre-incorporation, founder-attested)", "how_to_read": "Every claim carries exactly one status from the vocabulary below. Statuses are not collapsed into a generic 'verified' badge. A claim is upgraded only when the artefact named in supporting_artifacts exists, is public, and supports the exact text. INDEPENDENTLY_* requires a named external party; none exists for any CAIN-42 claim.", "status_definitions": { "DOCUMENTED": "Described in a document, but not demonstrated by a measurement.", "OPERATOR_MEASURED": "Measured by CAIN's operator. The measurement may be real and reproducible-in-principle but has not been repeated by an external party.", "REPRODUCIBLE": "Sufficient public materials exist for another party to repeat the relevant test in a clean environment.", "INDEPENDENTLY_REPRODUCED": "A named external party repeated the test and published results.", "INDEPENDENTLY_AUDITED": "A named auditor examined a stated scope and published findings.", "NOT_ESTABLISHED": "Required evidence is missing, or the run is incomplete, or the exact figure could not be located.", "FAILED": "The specified test or criterion failed.", "SUPERSEDED": "Replaced by a newer, linked result." }, "headline_claims": ["C42-CONSENSUS-4REPLICA", "C42-QC-3OF4-ED25519", "C42-FORMAL-7.3M-STATES", "C42-TESTS-1963-1978"], "claims": [ { "claim_id": "C42-CONSENSUS-4REPLICA", "claim_text": "Every hosted decision is ordered by a live 4-replica PBFT quorum.", "component": "PBFT cluster cain-mr-01 (and cain-mr-02)", "version": "image cain-cluster-node:mr-6047ff1, commit 6047ff1d2bc736538bed109ee8cf5f09c1d17174", "environment": "cainstudio.online production, 2026-10-09", "status": "OPERATOR_MEASURED", "supporting_artifacts": [ "https://cainstudio.online/api/v1/live-cluster/status", "https://cainstudio.online/api/v1/live-cluster/membership", "https://cainstudio.online/proof/bundle/hourly-proof/latest.json", "https://cainstudio.online/proof/bundle/hourly-proof/index.json", "https://cainstudio.online/evidence/artifacts/deploy-cain-mr-01.redacted.json", "https://cainstudio.online/evidence/deployment.json" ], "reproduction": [ "curl -s https://cainstudio.online/api/v1/live-cluster/status", "curl -s https://cainstudio.online/api/v1/live-cluster/membership", "Compare the returned membership public keys to /.well-known/cain-keys.json." ], "assumptions": [ "The four replicas are addressed over a private WireGuard overlay (10.42.0.0/24) that the gateway, not the reviewer, reaches.", "cain-mr-01 runs on 3 physical hosts (atl, lax, lax, mia) on a single provider (Vultr); '4 replicas' does not mean 4 independent failure domains." ], "limitations": [ "At 2026-10-09T16:10Z the live status reported agreement.same_height=false, agreement.same_state=false, same_view=false, node-2 quarantined, node-1 commit_index 47377 vs 47277 on the others, quorum_state DEGRADED.", "The latest hourly proof (proof-000533) returned verdict NOT_OPERATIONAL: 'write not committed: Primary unreachable; view-change initiated'.", "The historical claim '4 replicas on 4 servers across 4 regions' is FALSE for cain-mr-01 (3 regions, 3 hosts). It is true for cain-mr-02 (atl/lax/mia/sjc, one replica each), still on one provider. The '4 servers / 4 regions' text is corrected by C42-CONSENSUS-4REGIONS-CORRECTION.", "Operator-attested placement: a redacted manifest does not independently prove physical or geographic placement." ], "last_checked_utc": "2026-10-09T16:10:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": [ "A live /api/v1/live-cluster/status returning fewer than 4 configured replicas or a membership with n != 4.", "Membership public keys that do not match the pinned keys.", "Two replicas with two different configuration hashes." ] }, { "claim_id": "C42-CONSENSUS-4REGIONS-CORRECTION", "claim_text": "CORRECTION: cain-mr-01 is 4 replicas on 3 hosts across 3 regions (atl, lax, lax, mia) on one provider; cain-mr-02 is 4 replicas on 4 hosts across 4 regions (atl, lax, mia, sjc) on one provider. No CAIN-42 cluster currently spans more than one infrastructure provider.", "component": "deployment", "version": "deploy records 2026-10-05", "environment": "production", "status": "OPERATOR_MEASURED", "supporting_artifacts": [ "https://cainstudio.online/evidence/artifacts/deploy-cain-mr-01.redacted.json", "https://cainstudio.online/evidence/artifacts/deploy-cain-mr-02.redacted.json", "https://cainstudio.online/proof/bundle/hourly-proof/index.json" ], "reproduction": ["Fetch the two redacted deploy records; group replicas by region and by host."], "assumptions": ["Regions and hosts are read from the operator's redacted deploy records."], "limitations": ["Public host IPs are redacted, so geographic placement is not independently verifiable; provider diversity is absent."], "last_checked_utc": "2026-10-09T16:10:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A deployment where two replicas resolve to the same public host.", "A cluster with replicas on more than one provider."] }, { "claim_id": "C42-QC-3OF4-ED25519", "claim_text": "A 4-node CAIN-42 PBFT cluster produces authentic quorum certificates (>= 3 of 4 pinned Ed25519 members) with an identical decision chain on every node.", "component": "cain_pbft_engine_33 / PBFT QC bundle", "version": "engine version 33.4-production-hardened (bundle)", "environment": "disposable 4-node cluster (bundle); live clusters cain-mr-01/02 (operator-attested)", "status": "REPRODUCIBLE", "supporting_artifacts": [ "https://cainstudio.online/proof/bundle/pbft-evolution2-2026-09-24/PBFT_QC_BUNDLE.json", "https://cainstudio.online/proof/bundle/pbft-evolution2-2026-09-24/verify_pbft_qc_bundle.py.txt", "https://cainstudio.online/proof/bundle/pbft-evolution2-2026-09-24/VERIFIER_OUTPUT.txt", "https://cainstudio.online/evidence/verification-pack/verify_pbft_qc_bundle.py", "https://cainstudio.online/evidence/verification-pack/PBFT_QC_BUNDLE.json" ], "reproduction": [ "python3 -m pip install cryptography", "python3 verify_pbft_qc_bundle.py PBFT_QC_BUNDLE.json", "Expected: ALL_CHECKS_PASSED: 29/29 checks, 160 certificates verified (verified 2026-10-09 by the operator on Python 3.14.4 / cryptography 46.0.5)." ], "assumptions": ["The pinned Ed25519 member keys are the trust root.", "The bundle was produced by a disposable single-host cluster, not the live cain-mr-01/02."], "limitations": [ "Signatures prove that the named keys signed the named bytes; they do not prove independent machines, distributed consensus, or infrastructure independence.", "The live certificate endpoint /api/v1/live-cluster/qc/{seq} timed out during the 2026-10-09 check; live QC retrieval is OPERATOR_MEASURED, not independently reproduced.", "Static trust root: the verifier cannot observe live revocation or replay of a certificate it is not told about." ], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": [ "The verifier rejecting the published bundle.", "A published bundle with a COMMIT_QC signed by fewer than 3 distinct members.", "Two nodes in the bundle with different decision chains." ] }, { "claim_id": "C42-FORMAL-7.3M-STATES", "claim_text": "TLA+ models of the PBFT commit/view-change rules and the MCPGate authorization gate were checked by TLC: 0 violations over approximately 7.3 million distinct states, and 8 deliberately broken variants were each caught.", "component": "docs/formal-verification + scripts/formal", "version": "run 2026-09-27", "environment": "operator host, TLC (tla2tools.jar), Java 11+", "status": "REPRODUCIBLE", "supporting_artifacts": [ "https://cainstudio.online/proof/bundle/formal-2026-09-27/CAINPBFTCommitSafety.tla.txt", "https://cainstudio.online/proof/bundle/formal-2026-09-27/CAINMCPGateAuthorization.tla.txt", "https://cainstudio.online/proof/bundle/formal-2026-09-27/CAINPBFTCommitSafety.tlc.json", "https://cainstudio.online/proof/bundle/formal-2026-09-27/CAINMCPGateAuthorization.tlc.json", "https://cainstudio.online/proof/bundle/formal-2026-09-27/REPRODUCE.txt" ], "reproduction": [ "Follow REPRODUCE.txt: download the .tla.txt files as .tla, write the .cfg from the file, run java -cp tla2tools.jar tlc2.TLC -deadlock -config .", "The aggregate is the sum of 5 successful runs: PBFT 290,806 + 670,340 + 4,198,642; MCPGate 553,996 + 1,591,308 = 7,305,092 distinct states." ], "assumptions": [ "7.3M is a SUM of DISTINCT states across 5 TLC configurations (3 PBFT + 2 MCPGate), not one model.", "Bounds: PBFT N=4, f<=1, Q=3, one sequence, views {0,1}, values {X,Y}; MCPGate N=4, Q=3, 2 proposals/ids/actions/actors/contexts, time 0..2." ], "limitations": [ "These are bounded, abstract models. They are NOT a proof about the Python implementation, and no proved refinement exists between model and code.", "Liveness is not checked. Symmetry reduction: not used. The '0 violations' applies only to the two safety invariants per model.", "The 7.3M figure is a rounded label for 7,305,092 distinct states; per-run figures are in the .tlc.json files." ], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": [ "Re-running TLC and finding a violated invariant.", "A mutant that the run fails to catch.", "Per-run distinct-state counts that do not sum to the published total." ] }, { "claim_id": "C42-TESTS-1963-1978", "claim_text": "A full test run collected 1,978 tests: 1,963 passed, 0 failed, 0 errors, 7 skipped, 8 expected failures.", "component": "CAIN-42 defined suites (site, gateway, hypervisor, L5 governance, MCP enforcement, Byzantine)", "version": "git commit 068b79839a752c633b75dc2a3fe3411fce6724c4", "environment": "operator host, Python 3.14.4, Linux x86_64; NOT independent CI", "status": "OPERATOR_MEASURED", "supporting_artifacts": [ "https://cainstudio.online/proof/bundle/proof-fabric/TEST_MANIFEST.json", "https://cainstudio.online/proof/bundle/proof-fabric/junit/site.xml", "https://cainstudio.online/proof/bundle/proof-fabric/junit/gateway.xml", "https://cainstudio.online/proof/bundle/proof-fabric/junit/hypervisor.xml", "https://cainstudio.online/proof/bundle/proof-fabric/junit/l5-governance.xml", "https://cainstudio.online/proof/bundle/proof-fabric/junit/mcp-enforcement.xml", "https://cainstudio.online/proof/bundle/proof-fabric/junit/byzantine.xml" ], "reproduction": [ "The per-suite commands are recorded in TEST_MANIFEST.json.", "NOT FULLY REPRODUCIBLE PUBLICLY: the manifest states 'the suites need the repository (source not public by operator directive)' and no public git remote exists. A reviewer without the source can read the manifest and JUnit XML but cannot re-run the suite." ], "assumptions": ["The 1,978 figure counts collected tests across the six named suites only, not all 10,007 test functions in the repository."], "limitations": [ "Arithmetic reconciles internally: 1,963 passed + 7 skipped + 8 expected failures = 1,978; failed 0, errors 0. The 'remaining 15' are exactly the 7 skips and 8 expected failures.", "A whole-repository run is not the same scope; the 6 suites are a subset.", "The operator's own 2026-10-02 audit could not reproduce the full run; a 23-test slice gave 20 pass / 3 fail against a moved artefact." ], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["An external re-run yielding a different pass/fail total.", "A test that passes in the report but fails on the same commit."] }, { "claim_id": "C42-TESTS-1857", "claim_text": "An earlier run of the same six suites collected 1,857 tests: 1,842 passed, 0 failed, 7 skipped, 8 expected failures.", "component": "same suites", "version": "commit not published (no public git remote); recorded in artifacts/e42", "environment": "operator host", "status": "SUPERSEDED", "supporting_artifacts": ["https://cainstudio.online/proof/bundle/proof-fabric/TEST_MANIFEST.json"], "reproduction": ["Reconcile with C42-TESTS-1963-1978: 1,842 + 7 + 8 = 1,857."], "assumptions": [], "limitations": ["Superseded by the 1,978-test run at commit 068b7983 on 2026-09-28."], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A run at the same commit giving a different total."] }, { "claim_id": "C42-TESTS-870", "claim_text": "870 tests passed, 0 failed, across the consensus, pipeline, site and parity suites.", "component": "subset of suites", "version": "commit 9baead0e897e4ce7b92cac11ccd69d6a878d1b2a", "environment": "operator host, 2026-09-26T22:15:27Z", "status": "SUPERSEDED", "supporting_artifacts": ["https://clawx.click/evidence/test-report-2026-09-26/TEST_REPORT.json"], "reproduction": ["The report is at the URL above (clawx.click is a separate CAIN site)."], "assumptions": [], "limitations": ["Different date, commit and scope from the 1,978 run; the two totals are not comparable."], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A re-run at commit 9baead0e giving a different count."] }, { "claim_id": "C42-QC-ZERO-SIGNATURE-INCIDENT", "claim_text": "Historical defect (now fixed): an action-proof verifier reported a valid proof while accepting placeholder signatures (a count check, len(proof_signatures) >= 3, not a cryptographic check), and a separate clean-room verifier accepted a bundle with threshold 0 and no signatures. Both accepted zero validating certificate signatures.", "component": "platform-gateway/cain_action_proof_epoch6.py (#83); verification/govbpf/verify_e13.py (#44)", "version": "pre-fix", "environment": "offline proof/verifier paths", "status": "SUPERSEDED", "supporting_artifacts": [ "https://cainstudio.online/proof/bundle/top100-remediation-2026-10-07/REMEDIATION_LOG.md" ], "reproduction": [ "Read entries #83 and #44 of the remediation log.", "Regression tests: tests/test_cain42_epoch6_action_proof.py (3 added tests: real signature, forgery, cross-engine pinning); tests/govbpf/test_govbpf.py (9 added tests, incl. threshold validation)." ], "assumptions": ["These were proof/verifier path defects; the live MCPGate quorum uses a separate pinned-key verifier (cain_pbft_engine_33)."], "limitations": [ "Fix commit SHA and whether the hosted production decision path was affected are NOT_ESTABLISHED here: the repository has no public git remote, and this package did not re-run the historical exploit against the live host.", "The pre-fix verifier is intentionally preserved in the published e13 bundle and superseded by the remediation bundle, so evidence is not rewritten." ], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["Re-running the current verifier on a crafted zero-signature bundle and getting a PASS."] }, { "claim_id": "C42-ENFORCE-MODE", "claim_text": "The hosted CAIN Studio enforcement Fabric runs in enforce mode: policy, authorization, risk, verification and consensus stages block on the request path.", "component": "platform-gateway /fabric", "version": "live 2026-10-09", "environment": "cainstudio.online", "status": "OPERATOR_MEASURED", "supporting_artifacts": ["https://cainstudio.online/fabric/status"], "reproduction": ["curl -s https://cainstudio.online/fabric/status and read .mode and .enforcement.stages."], "assumptions": [], "limitations": [ "The same response shows some stages NOT enforcing: intent (off by default), approval (off by default), evidence (retention only). 'Enforce mode' is the master switch, not every stage.", "Enforcement depends on the caller routing its tool call through the gate; calls that bypass the gate are not governed (see C42-MCPGATE-BOUNDARY)." ], "last_checked_utc": "2026-10-09T16:10:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["/fabric/status reporting mode != 'enforce'."] }, { "claim_id": "C42-MCPGATE-BOUNDARY", "claim_text": "A tool call routed through MCPGate runs only with a quorum-committed authorization; calls that bypass the gate are not governed by it.", "component": "MCPGate / cain/mcp_proxy", "version": "live 2026-10-09", "environment": "cainstudio.online (mcpgate deployment)", "status": "OPERATOR_MEASURED", "supporting_artifacts": [ "https://cainstudio.online/fabric/status", "https://cainstudio.online/evidence/known-limitations.md" ], "reproduction": ["Read .enforcement.stages.mcpgate and .consensus in /fabric/status."], "assumptions": ["The protected tool must be reachable only through the gate for enforcement to be meaningful."], "limitations": [ "The boundary is opt-in: an agent that can reach an equivalent tool outside the gate is not governed by CAIN.", "The full bypass test matrix (allowed succeeds, denied does not reach the tool, parameter change rejected, replay rejected, tenant isolation, delegation attenuation, restart durability, fail-open/fail-closed) has NOT been publicly reproduced from outside; those are NOT_ESTABLISHED in this package (see C42-ENFORCEMENT-DEMOS)." ], "last_checked_utc": "2026-10-09T16:10:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A denied action reaching the protected tool.", "A tool call entering the gate without an authorization."] }, { "claim_id": "C42-ENFORCEMENT-DEMOS", "claim_text": "Ten controlled bypass-resistance demonstrations (allowed succeeds; denied does not reach the tool; changed parameters rejected; direct access blocked; replay rejected; revoked/expired authority rejected; cross-tenant authority rejected; delegation cannot widen; restart preserves replay protection; dependency/verifier failure follows documented fail-closed behaviour).", "component": "enforcement boundary", "version": "n/a", "environment": "authorized test environment", "status": "NOT_ESTABLISHED", "supporting_artifacts": [], "reproduction": ["No public, reproducible, side-effect-showing demonstration exists in this package."], "assumptions": [], "limitations": [ "A side-effecting bypass test requires explicit operator approval and will not be run against production or customers.", "Prior self-run bundles (CAIN_20_BYPASS_MATRIX.json, CAIN42_L5_BYPASS_REPORT.json) exist in the source tree but were not re-run and are not published as public artefacts here." ], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["Any of the ten tests producing the wrong outcome in an authorized environment."] }, { "claim_id": "C42-SOAK-72H", "claim_text": "A 72-hour multi-region soak of cain-mr-01 passed with 0 divergences and 0 safety violations.", "component": "multi_region_soak.py, cain-mr-01", "version": "harness in run", "environment": "Vultr, 3 hosts, 3 regions", "status": "NOT_ESTABLISHED", "supporting_artifacts": [ "https://cainstudio.online/proof/bundle/soak-multiregion-2026-10-06/heartbeat.json", "https://cainstudio.online/proof/bundle/soak-multiregion-2026-10-06/index.json" ], "reproduction": ["Read the heartbeat: state RUNNING, elapsed_hours, divergences, anomalies."], "assumptions": [], "limitations": [ "The current run is not finished: at 2026-10-09T16:11Z it reported RUNNING, 34.1h of 72h, 0 divergences, 0 anomalies, 95 injected faults / 95 restarts.", "The evidence directory is named soak-multiregion-2026-10-06 but its published heartbeat started_at is 2026-10-08T06:06:05Z; an earlier run segment started 2026-10-06T01:36:59Z (see C42-SOAK-OCT6-RUN).", "Preceding runs did not pass: soak-multiregion-2026-09-26 failed at checkpoint 32; soak-multiregion-2026-09-29 and -29b and soak-120h-mr02-2026-09-29 were STOPPED_BY_OPERATOR." ], "last_checked_utc": "2026-10-09T16:11:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A heartbeat showing a divergence or anomaly > 0.", "A completed 72h run that fails its predeclared criteria."] }, { "claim_id": "C42-SOAK-OCT6-RUN", "claim_text": "Resolution of the run that started 2026-10-06T01:36:59Z: it did NOT complete a published 72h run. It ran and produced hourly checkpoints under the same name/working directory, and the currently published soak-multiregion-2026-10-06 evidence is a later segment that started 2026-10-08T06:06:05Z. The later segment is still RUNNING (34.1h of 72h at 2026-10-09T16:11Z).", "component": "cain-mr-soak.service / multi_region_soak.py", "version": "service definition names the run 'soak-multiregion-2026-10-06'", "environment": "cain-mr-01", "status": "OPERATOR_MEASURED", "supporting_artifacts": [ "https://cainstudio.online/proof/bundle/soak-multiregion-2026-10-06/heartbeat.json", "https://cainstudio.online/proof/bundle/soak-multiregion-2026-10-06/REPRODUCE.txt" ], "reproduction": ["Compare the heartbeat started_at with the checkpoint-0000 timestamp; both are 2026-10-08."], "assumptions": ["The operator host log (/root/cain-mr-soak-2026-10-06/soak.log, first line) records the 2026-10-06T01:36:59Z start; it is not a public artefact."], "limitations": [ "Predeclared criteria: 72.0h planned, fault every 1200s, hourly hash-chained checkpoints, divergences == 0. No evidence that criteria were changed after the run started; the harness definition fixes 72h.", "The Oct-06 segment's published checkpoints were overwritten by the Oct-08 segment (same directory), so the Oct-06 segment cannot be independently verified from the published bundle." ], "last_checked_utc": "2026-10-09T16:11:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A published Oct-06 checkpoint set with an Oct-06 start timestamp."] }, { "claim_id": "C42-PERF-SNAPSHOT", "claim_text": "Consensus latency of approximately 2,269 ms per fresh round is reduced to approximately 0.83 ms on the quorum-committed authorization-snapshot path.", "component": "authorization snapshot / governance data plane", "version": "internal register", "environment": "operator measurement", "status": "OPERATOR_MEASURED", "supporting_artifacts": ["https://cainstudio.online/proof/bundle/proof-fabric/PERFORMANCE_INDEX.json"], "reproduction": ["The originating measurement is an operator document; PERFORMANCE_INDEX.json marks the published benchmarks as INCOMPLETE conditions."], "assumptions": ["The two figures measure different boundaries: fresh consensus vs cached/snapshot authorization."], "limitations": [ "No public benchmark code, hardware spec, concurrency, sample count or percentiles accompany these exact figures.", "They must not be presented as interchangeable latency improvements." ], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A public, reproducible benchmark showing different numbers at the same boundary."] }, { "claim_id": "C42-PERF-SUB15MS", "claim_text": "sub-15ms verification and a browser demonstration below 2ms.", "component": "gateway decision path / browser demo", "version": "n/a", "environment": "marketing and demo material", "status": "NOT_ESTABLISHED", "supporting_artifacts": ["https://cainstudio.online/proof/bundle/proof-fabric/PERFORMANCE_INDEX.json"], "reproduction": ["No public benchmark code matching these statements was located."], "assumptions": [], "limitations": ["The performance index marks the published benchmarks incomplete; the exact measurement boundary for 'sub-15ms' and 'below 2ms' is undefined in the public material."], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A public benchmark at the stated boundary exceeding the stated threshold."] }, { "claim_id": "C42-REVENUE-FIRST", "claim_text": "CAIN-42 / FIRME AI has its first paying customers (founder-attested: 2 customers, first revenue $96 + $1 = $97).", "component": "company", "version": "founder attestation 2026-10-12 (as recorded in AGENTS.md)", "environment": "commercial", "status": "NOT_ESTABLISHED", "supporting_artifacts": [], "reproduction": ["No public processor/invoice record is published."], "assumptions": [], "limitations": [ "Evidence class is FOUNDER_ATTESTED_UNVERIFIED; it is not ARR, recurring revenue, or enterprise traction.", "No customer reference or payment-processor record is published." ], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A published processor record that contradicts the attestation."] }, { "claim_id": "C42-INDEPENDENT", "claim_text": "Any claim of independent external verification, reproduction, or audit.", "component": "whole platform", "version": "n/a", "environment": "n/a", "status": "NOT_ESTABLISHED", "supporting_artifacts": [], "reproduction": ["No named external party has reproduced or audited any CAIN-42 claim."], "assumptions": [], "limitations": ["All evidence in this package is OPERATOR_GENERATED. Independent verification is 0."], "last_checked_utc": "2026-10-09T16:05:00Z", "responsible_publisher": "CAIN-42 operator", "falsifiers": ["A named external party publishing a reproduction or audit."] } ] }