You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Security: Omkar-git-hub/msk-project-intelligence
Security
docs/security.md
MSK Security Model
1. Principles
Zero Secret Persistence: Under no circumstance are secret keys, passwords, bearer tokens, private keys, or credentials stored inside .msk/msk.db or .msk/*.json.
Local Analysis Only: Core analysis operates offline without external network communication.
Boundary Integrity: In future AI phases, context passed to LLMs must pass through strict policy filtering and redaction.
Symlink Traversal Protection: The file scanner rejects and skips symlinks that point outside the project boundary or create directory recursion loops.
Safe Subprocess Execution: Any interaction with native Git commands uses parameterized lists rather than raw shell strings to prevent injection.