Real Captures. Real Data.
Both events below were captured live by the C-Engine triangulation mesh. No simulations, no reconstructed data. The raw JSON payloads are exactly what a connected algorithm received at the moment of detection.
RBF Fee Replacement — Live Interception
A 14.47 BTC transaction was intercepted at the P2P layer during an active fee replacement sequence. The sender had already broadcast a version at 128 sats/vB and bumped it to 151 sats/vB — an 18% fee increase on roughly $1.08M of value. The engine detected the replacement within milliseconds of the newinvmessage hitting the mesh and fired the RBF_PANIC classification before the replacement propagated to standard nodes.
Raw Telemetry — Received at T+0
{
"asset": "BTC",
"type": "RBF_PANIC",
"size_btc": 14.4765,
"txid": "34a524eee0474b9471f1f38103b28b6eff0086191ccab03221212bd90527dd95",
"rbf_enabled": true,
"rbf_bump_count": 1,
"node_ip": "100.0.184.176",
"shock_score": 0.11,
"fee_velocity": 0.7,
"rbf_aggression": 0.2,
"microburst": false,
"network_void": false,
"pressure": { "0-20": 0.67, "20-50": 0.24, "50-100": 0.04, "100-200": 0.02, "200+": 0.03 },
"fee_acceleration_curve": [128.0, 151.0]
}Why this TXID doesn't appear on a block explorer
RBF replacement transactions are by design ephemeral. Once a transaction is replaced by a higher-fee version, the original TXID is evicted from the mempool and never confirmed into a block. Block explorers like mempool.space only index the final confirmed version — the original replacement TXID disappears permanently. This is not a limitation of the engine. It's proof the engine caught something that no after-the-fact lookup tool can retrieve. The interception window between broadcast and replacement is exactly what this product is built to exploit.
Legacy Whale — Independently Verifiable
A 6.39 BTC legacy-format transfer originating from node 66.92.204.11 was intercepted during a network void period — mempool pressure was suppressed, with 67% of activity in the sub-20 BTC range. The engine correctly identified the transaction as non-RBF with no fee replacement risk, classifying it as a clean institutional sweep. Because this is a legacy transaction (no SegWit witness data), the TXID the engine computes is identical to the on-chain TXID. You can verify it independently on any block explorer.
Raw Telemetry — Received at T+0
{
"asset": "BTC",
"type": "LIVE_LEGACY_WHALE",
"size_btc": 6.3915,
"txid": "bd110c7f4221c359ff5cac5a26ec090c73e8308cf55066cd7ef491c15a5dc969",
"rbf_enabled": false,
"rbf_bump_count": 0,
"node_ip": "66.92.204.11",
"shock_score": 0.09,
"fee_velocity": 0.0,
"rbf_aggression": 0.0,
"microburst": false,
"network_void": true,
"pressure": { "0-20": 0.67, "20-50": 0.33, "50-100": 0.0, "100-200": 0.0, "200+": 0.0 },
"fee_acceleration_curve": [125.0]
}Independently verifiable on-chain
Legacy transactions carry no witness data, so the TXID computed by the engine is identical to the TXID confirmed on-chain. Paste the hash below into any block explorer and you'll find the exact transaction the engine intercepted. This is what independent verification looks like for non-RBF institutional transfers.
mempool.space/tx/bd110c7f4221c359ff5cac5a26ec090c73e8308cf55066cd7ef491c15a5dc969 →A note on shock scores
Both captures above returned low shock scores (0.09–0.11). That's intentional and honest — most mempool activity is routine. The shock score is a composite signal designed to spike above 0.80 only during genuine coordinated fee market pressure: sustained RBF sequences, microburst density above 3 whales/second, or high fee velocity compounding across a short window. What these captures demonstrate is not maximum shock — it's that the engine is reading the live network accurately and classifying events correctly in real time. A 0.09 during a network void and a 0.11 during a single-bump RBF are exactly the right outputs. If everything scored 0.94, the signal would be noise.