elephant-diffuser implements the BitLocker Elephant Diffuser — a keyed byte-mixing transform applied to untrusted sector bytes from BitLocker (BDE) volumes, including images acquired from compromised or actively hostile systems. Hostile input is the expected case. Robustness against arbitrary sector content and length is a core design goal, and we take reports of crashes, hangs, or memory-safety issues seriously.

Do not open a public GitHub issue for a security vulnerability.

Report privately, by either:

• GitHub Security Advisories — open a private advisory on the elephant-diffuser repository, or
• Email — albert@securityronin.com.

Please include:

• the affected version and target triple,
• a minimal reproducing input (a fuzz corpus entry is ideal),
• the observed behaviour (panic, hang, excessive allocation, wrong output) and the expected behaviour.

We aim to acknowledge a report within a few business days and to coordinate disclosure once a fix is available.

• #![forbid(unsafe_code)] — no unsafe, anywhere.
• No panics on any input — the transform runs on a sector of any length, including empty and sub-word buffers; every index is computed with underflow-free modular arithmetic, so no length can drive an out-of-bounds access or an arithmetic overflow.
• No unwrap/expect in production — clippy::unwrap_used/expect_used are denied outside tests.
• Continuous fuzzing with cargo-fuzz: decrypt/encrypt over arbitrary bytes, invariant "must not panic," smoke-run on every push/PR and fuzzed for ten minutes weekly.

The Elephant Diffuser holds no secret and branches on no secret; it is a diffusion layer, not a confidentiality cipher. elephant-diffuser is correctness-validated against libbde on real data (see docs/validation.md), not independently security-audited.