This work presents a novel manifest-based fingerprinting approach for detecting malicious browser extensions in the Chrome ecosystem. Browser extensions have become a critical attack surface—prior studies estimate that roughly 10% of published extensions exhibit security noteworthy or outright malicious behavior. At the same time, Google’s migration from Manifest V2 to V3 (MV2 to MV3) introduces both new security controls and conversion challenges that impact detection strategies. First, we provide the most up-to-date measurement of MV3 adoptionin the Chrome Web Store (CWS), finding that only 65.83% ofextensions had migrated to MV3 as of January 25, 2025—leaving asubstantial legacy MV2 population only months before MV2 deprecation. We then evaluate the real-world usage of GoogleChromeLabs’ Extension Manifest Converter (EMC), showing that under 1.2% of MV3 extensions show its conversion artifacts, indicating the tool's negligible adoption despite its promise. Building on these findings, we develop a four-stage pipeline to extract, compare, and weight manifest-change tokens from known malicious MV2 to MV3 conversions. Although EMC-derived patterns alone proved too sparse for reliable detection, we adapted our methodology to mine MV3-native malicious extensions and derive a granular scoring function over manifest tokens. Applied to 88,327 MV3 extensionsin the CWS, our fingerprint scores closely match prior estimatesof security-noteworthy extension prevalence (approx. 12.8% above a0.5 threshold) and distinguish malicious samples with 85.32% scoringabove 0.5. Our work (1) delivers an empirical lens on MV3 adoption and EMC usage, (2) introduces a lightweight, manifest-only detection mechanism that requires no code analysis or dynamic instrumentation, and (3) charts a new path for maintaining extension security in the post-MV2 era.