The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides documentation and templates for utilizing the platform's native run_command tool and event hook system. These are documented as standard extension points for users to automate workflows (e.g., running linters or security checks) and do not contain pre-configured malicious commands.
[EXTERNAL_DOWNLOADS]: The documentation includes examples for configuring Model Context Protocol (MCP) servers using npx and uvx. These examples reference official and well-known packages such as @modelcontextprotocol/server-github and @modelcontextprotocol/server-filesystem, which are standard components within the ecosystem.
[CREDENTIALS_UNSAFE]: The skill documents how to pass authentication tokens to MCP servers via environment variables. It correctly uses non-sensitive placeholders such as <YOUR_TOKEN> and <YOUR_PERSONAL_ACCESS_TOKEN> in its examples, ensuring no actual credentials are exposed while instructing users on proper configuration.

agy-extensions