Add ilang-ai skills (everything-is-ok, less-token, no-prompt)

[adsorgcn]

Three ClawHub-verified skills (Benign HIGH CONFIDENCE), all instruction-only, MIT licensed:

• everything-is-ok — Universal AI upgrade protocol. Improves any LLM's output structure and token efficiency.
• less-token — 40-65% token reduction on summarization. Zero dependencies.
• no-prompt — Prompt engineering eliminator. AI-to-AI instruction generation.

Built on I-Lang Protocol. GitHub: https://github.com/ilang-ai/ilang-openclaw

[Anil-matcha]

Please add a PR

[adsorgcn]

Sure, I'll submit a PR shortly. Thanks for the guidance!