"author:SoWhatI" matching MCP servers:

Enables running shell commands and persistent sessions with secrets stored in local secure storage injected into child processes at exec time based on rules. Guarantees secret plaintext never reaches the model context, returning only names, masks, and redacted output.

Enables AI clients to securely access and control diverse backend assets through a unified MCP entry point, with centralized management of credentials, authorization groups, and audit logging.