The assignment
Create your personal voice skill — one that communicates in your style. Pull from LinkedIn posts and writing samples.

MyVoiceSkill, an AI skill published on GitHub as niraj-linkedin-voice, with a SKILL.md, README, and a references folder. It's built around a specific structural pattern the author identified in his own posts - setting up an assumption, then breaking it - checked against a five-point validation list before finalizing a draft.
Niraj shipped a public GitHub repo. This read comes from the artifact and Niraj's own account of it — no scores, just what the build does and does not show.
What it demonstrates
What is not shown
Take this awayThe reusable lesson: a voice is a file, not a prompt. Named, described, fed on your real posts, it means every future draft starts closer to you.
A reusable Skill beats a clever one-off prompt every time. Packaging your voice as a named, described, instructed artifact is the difference between a trick you re-type and an asset you — and your team — can run tomorrow.
Seeded on writing that is actually yours
Real LinkedIn posts and writing samples as source material. A voice built from adjectives — 'punchy, warm, contrarian' — sounds like everyone else's.
Packaged as an asset, not a prompt
A name, a description, instructions, a file someone else could run tomorrow. That is the difference between a skill and a paragraph you re-type.
A voice match a reader can judge
The skill's output shown next to something the builder genuinely wrote, so the resemblance is visible rather than asserted.
No numeric scores are published here. These pages are written reads of what each builder shipped — the build is the point, not a mark out of a hundred.