"Shell" matching MCP connectors:

Ask if an action is safe before your agent runs it: injection, shell, MCP installs. Free samples.

Deterministic security preflight for AI agents. Check URLs, files and shell commands before acting.

Diff two npm/PyPI versions or lockfiles: new install scripts, network/shell/eval calls, new deps.

Audit Dockerfiles for root users, baked-in secrets, curl-pipe-shell and unpinned base images.