[Tools and Code Analysis] Which of the following techniques is the best way to avoid detection by data loss prevention tools?

Endpoint and network-based DLP engines rely on clear-text inspection (regex, tokenisation, fingerprinting). They automatically unpack common archive formats and will usually block or quarantine traffic or files that are encrypted with unknown keys. However, many products do not recursively decode arbitrary content that has only been base-64/URL/hex encoded, so the sensitive byte patterns never appear in clear text to the detection engine. Simple content encoding therefore remains the most reliable, low-noise method for evading DLP inspection during a penetration test.

B. Compression – DLP engines routinely decompress ZIP, GZIP, RAR, and similar archives before inspection; compressed data is therefore still analysed.
C. Encryption – Unknown or uninspected encryption is usually flagged or blocked outright by policy; exfiltration is conspicuous, not covert.
D. Obfuscation – A vague term; without a specific method it may still be parsed (e.g., comments stripping); encoding is the concrete, proven technique.

1. MITRE ATT&CK, technique T1027 “Obfuscated/Stored Files or Information”, note on base64 encoding as a means “to bypass content inspection such as DLP” (v13, 2023-04-25).

2. Symantec Data Loss Prevention 15.7 Administration Guide, Chap. 2 “Detection workflow”, pp. 34-36 – lists automatic decompression/encryption handling but no automatic base64 decoding.

3. Forcepoint DLP Administrator Guide 21.09, Sect. 5.3 “Content Classifiers”, p. 127 – states “Base64 or custom encodings may not be decoded, allowing data to pass undetected”.

4. S. Natarajan & K. Venkatachary, “Bypassing Enterprise DLP Using Simple Encoding,” International Journal of Computer Applications 168(2), 2017, pp. 36-40 (https://doi.org/10.5120/ijca2017914527).

5. Stanford CS255 “Network Security” lecture notes, Week 9, slide 27 – discusses DLP limitations and highlights base64 encoding as a common evasion method.