Security audit
Security checks for vulnerabilities and agentic risk
The skill is rated suspicious because it uses an unpinned `npx groundmail-cli` command, which could run an unreviewed package version; the scan evidence is incomplete.
Review the skill before installing. Consider whether you are comfortable with its unpinned CLI commands; the scanner recommends using a reviewed, explicitly versioned CLI or a preinstalled dependency with verified integrity. The scan was partial, so the results do not establish that the skill is safe.
npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.
Although this invokes a CLI rather than an MCP server, npx groundmail-cli can download and execute a package without selecting a reviewed version. A compromised package release could execute with the agent's local permissions and access environment credentials; the skill's mailbox-scoped key guidance reduces API exposure but does not sandbox local execution.
The owner-registration example executes an unversioned npm CLI, not an MCP server. If package resolution selects a compromised release, it could steal the mailbox key or owner information and execute arbitrary code with the agent's permissions.
The verification example again invokes an unpinned CLI package, allowing execution of a release that has not necessarily been reviewed. A compromised release would receive the verification code and could access available credentials and local resources, making the authentication context particularly sensitive.
The workspace-login instruction uses an unversioned npm CLI rather than an MCP server. A compromised release could intercept login credentials or tokens and execute with local permissions; workspace authentication may expose broader access than the recommended mailbox-scoped key.
