In the digital era, organizations routinely process large volumes of confidential information, including personal identifiers, financial credentials, and healthcare data. Safeguarding this information without compromising its analytical value has become a crucial challenge. Conventional anonymization methods such as redaction or suppression often reduce data utility or fail to guarantee sufficient privacy. This paper presents a Policy-Driven Smart Data Anonymization Framework integrated with a Secure Credential Vault that ensures both privacy protection and operational usability. The proposed framework applies flexible masking strategies—full, partial, format-preserving, and noise-injection—to protect structured and unstructured datasets. It features a configurable policy engine for defining reusable anonymization rules adaptable across multiple business domains. An integrated PII Detection Module employs pattern recognition and context-aware text analysis to identify sensitive data such as emails, phone numbers, IP addresses, and financial details. The Credential Vault, secured using AES-256 encryption, manages passwords and API keys with encrypted backups and password-strength validation. Designed for offline use, the system supports CSV, Excel, JSON, and log files, offering real-time anonymization previews and compliance-ready audit reports. The overall solution demonstrates an effective balance between data confidentiality, usability, and regulatory compliance for diverse organizational and academic environments