Consider licensing skills under MIT-0 instead of MIT

Would you consider licensing the skill files under **MIT-0** instead of the regular MIT license?

The reason is that AI skills are often copied directly into another repository and then loaded into an agent's context. Under the MIT license, copies or substantial portions of the work need to retain the copyright and permission notice.

For normal software dependencies this is usually not much of an issue, since licenses can live alongside the dependency without affecting runtime behavior. For AI skills, however, additional license files, headers, or notices may end up being discovered or loaded by the agent, adding context/token overhead without providing any technical value.

MIT-0 seems particularly well suited to reusable AI skills:

* It retains the permissive nature of MIT.
* Users can copy and modify skills freely.
* There is no requirement to carry the copyright/license notice into downstream repositories.
* Internal users can copy a skill into their own skill directory without adding licensing-related files or text.
* It avoids unnecessary non-functional content in agent context.

This would be especially useful for organizations that want to vendor and customize skills internally without distributing their modifications.

Would you be open to either:

1. changing the skill license from MIT to MIT-0, or
2. dual-licensing the skills under MIT / MIT-0?

No specific file or test is named. First inspect how this repository currently declares the MIT license and which skill files it covers, then confirm whether the maintainers choose MIT-0 or dual licensing. Done means the selected licensing terms are consistently reflected in the repository and its skill files.

Written by the indexing model from the issue text.