Your code never trains a model
AI agents review your code on AuditHub, and none of what they read is used to train a model.
AuditHub never trains a model on your code. For anything else about how your code is handled, ask us directly and you get an answer from the team.
The commitments below are the ones we state publicly. Your security team can ask for the rest.
AI agents review your code on AuditHub, and none of what they read is used to train a model.
The open source MCP server reads your organizations, projects and findings. Running a tool is a switch you turn on yourself, and it launches only Vanguard and OrCa.
See how AuditHub MCP worksWhere your code is processed, how long it is kept, who can access it and what security reviews cover it: tell us what your security team needs to know.
Contact UsAI agents review your code, and security tools bring the evidence.