Please report suspected vulnerabilities privately via GitHub's private vulnerability reporting on this repository (the "Security" tab, then "Report a vulnerability"). Please do not open public issues for security reports.

We aim to acknowledge reports within five business days.

raptorgo implements forward error correction, not authentication or confidentiality. By design:

• FEC repairs erasures; it does not authenticate packets. A decoder fed forged packets can reconstruct attacker-controlled bytes, which is why the public decoder requires an application-supplied verifier before delivery.
• Malformed or hostile inputs must never cause unbounded memory growth, panics, or non-termination. The Limits type bounds packet sizes, retained symbols, decoding attempts, and aggregate allocations; violations of those bounds are in scope as vulnerabilities.
• Bugs that cause the decoder to deliver bytes that were never verified, or to diverge from RFC 6330 in a way an attacker can exploit, are in scope.

Reports about applications that skip object verification are out of scope; that contract is documented and deliberate.