The Agent Skills Directory

[SAFE]: The skill's functionality is limited to reading and writing markdown files within the project's directory structure. It does not attempt to access sensitive system files or environment variables.
[SAFE]: There are no network requests to external domains. Example commands for testing are restricted to 'localhost' for development purposes.
[SAFE]: The instructions do not contain prompt injections, obfuscated code, or mechanisms for persistence or privilege escalation.
[SAFE]: The skill encourages secure development practices by requiring code-verifiable acceptance criteria and automated testing.

create-tasks