The Agent Skills Directory

[SAFE]: The skill provides instructions for using official vendor tools (cargo-ai) to perform diagnostics on the vendor's own platform. All commands and API interactions are scoped to the authenticated workspace and follow established security practices for CLI usage.
[COMMAND_EXECUTION]: The skill instructs the user to run various cargo-ai commands and SQL queries. These are legitimate administrative and diagnostic actions within the context of the platform's functionality (e.g., cargo-ai orchestration query execute, cargo-ai orchestration run get).
[EXTERNAL_DOWNLOADS]: The skill documents the installation of the official CLI package @cargo-ai/cli from the npm registry. As this is the vendor's own package and a standard installation method, it is considered safe and expected behavior.
[CREDENTIALS_UNSAFE]: The skill mentions authentication methods (--token, --email, --oauth) and the need for admin tokens for specific billing tasks. It correctly advises using environment variables for tokens in CI environments and does not contain hardcoded secrets.

cargo-diagnostics