Human-in-the-loop security layer. Intercepts high-risk commands and requires push notification approval.
Initial release of ClawShell: a human-in-the-loop security layer for shell commands. - Intercepts shell commands, analyzes risk, and requires push notification approval for high-risk actions. - Blocks critical commands automatically; allows medium/low-risk commands based on configurable rules. - Provides tools for secure shell command execution (`clawshell_bash`), status checks, and log retrieval. - Supports Pushover and Telegram for approval notifications. - All command decisions and approvals are logged for audit and debugging. - Highly configurable via environment variables and `config.yaml`, with support for custom risk rules.