SILENT-KILLER-2026-v1: Dev-Wallet Silence Analysis Datasets for Solana pump.fun Rug Prediction Study (Positive-Class n=52 and Control-Panel n=91)

SILENT-KILLER-2026-v1 is a reproducibility bundle for the paper "Dev-Wallet Silence Is Not a Rug Predictor on Solana Pump.fun: Empirical Evidence from a Positive-Class and Control-Panel Study" (Kamat, 2026). The bundle supports a null-result empirical finding: dev-wallet transaction silence (>=6 hours) in the 24-hour window before a rug does NOT distinguish rug tokens from non-rug tokens. Positive class (n=52 rug events) shows 96.15 percent silence rate (95 percent Wilson CI [87.0, 98.9]); control class (n=91 non-rug wallets) shows 98.90 percent silence rate (95 percent Wilson CI [94.0, 99.8]). Two-proportion z-test z=-1.10, p=0.27; Fisher's exact test p=0.30. Balanced accuracy 48.62 percent. Youden J index -2.75 percentage points. The silence signal fails as a rug predictor because non-rug dev wallets are also mostly silent in random 24-hour windows post-graduation. The finding is robust across silence-threshold sensitivity sweeps of {2, 4, 6, 12, 18, 24} hours. Observation window: rug event timestamps (positive class) span 2026-06-29 to 2026-07-17; control-class random 24-hour window end timestamps span 2026-06-15 to 2026-07-17. See manuscript Section 3 for full corpus construction details. Contents: silent_killer_pilot_events.jsonl (52 rows, positive class), rug_attacker_fingerprints.jsonl (26 unique attacker wallets), fpr_control_events.jsonl (91 rows, control class), silent_killer_pilot_report.json, fpr_control_report.json, tau_sensitivity_report.json, LICENSE.md, README.md, SCHEMA.md. Total bundle size 21.2 KB. SHA-256 checksum BD48B69555B637602AA4741235C663143B698EE18B44B8923F15B1DD0DF8C18A. Ethics: All data are public Solana base-58 addresses. No personally identifying information is included. Patent and commercial licensing disclosure: This dataset is released under CC-BY-4.0. The author holds pending USPTO Provisional Patent Applications in algorithmic decentralized-exchange trading and blockchain market microstructure, including UDAY P1 #64/022,461 (filed 2026-03-30, Micro Entity) and UDAY P2 #64/099,108 (filed 2026-06-25, Micro Entity). The release of this dataset under CC-BY-4.0 does not affect patent priority claims under 35 USC 102(b)(1)(A). Commercial re-implementation of specific methodologies covered by pending patent claims may require patent licensing. Academic re-implementation is unaffected. Version 1.1 (2026-08-06) — Documentation update, no data changes.- SCHEMA.md and README.md version-stamp bumped from 1.0 to 1.1- Citation DOI placeholder [TO_BE_ASSIGNED] replaced with concept DOI 10.5281/zenodo.21434629- Observation-window language clarified to match manuscript v1.3 (JOTE-adapted)- All six data files unchanged from v1.0; SHA-256 hashes identical For commercial licensing inquiries: [e-mail]