Write, draft, edit, or review blog posts for jaysoo.ca in Jack's voice. Enforces the personal blog style guide (voice, tone, structure, words to avoid, signature phrases). Use whenever drafting, revising, restructuring, or proofreading a blog post, or when Jack says "blog post", "write a post", "draft a post", "blog voice", "blog style", or points at a file under blog-ideas/.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
Write, draft, edit, or review blog posts for jaysoo.ca in Jack's voice. Enforces the personal blog style guide (voice, tone, structure, words to avoid, signature phrases). Use whenever drafting, revising, restructuring, or proofreading a blog post, or when Jack says "blog post", "write a post", "draft a post", "blog voice", "blog style", or points at a file under blog-ideas/.
Blog Writing (jaysoo.ca)
Write and edit blog posts in Jack's established voice. This skill is the entry point for any blog work; the authoritative rules live in the style guide.
Step 1 — Load the style guide (always)
Read the full guide before writing or editing any prose:
.ai/para/resources/blog-ideas/STYLE_GUIDE.md
(Inside the dot-ai-config repo itself, there is no .ai symlink — read dot_ai/para/resources/blog-ideas/STYLE_GUIDE.md.)
The guide is the source of truth. Everything below is a fast reference, not a replacement. If the guide and this file ever disagree, the guide wins — and update this file.
Step 2 — Draft or edit
Follow the guide's structure: Problem -> Existing Solutions -> Their Limitations -> Better Solution -> Benefits. Argue by demonstration (show it in code), not by authority. Acknowledge alternatives before advocating.
Code-example sequence, every time: (1) state what it demonstrates, (2) show it, (3) explain the takeaway. Never drop code without surrounding prose.
Step 3 — Pre-publish checklist
Run every draft against this before calling it done. Grep the draft for the banned words.
Voice: inclusive "we" by default. "I" only for series intros / flagged opinions / disagreements. No "dear reader", "imagine you're...", "you might be thinking...".
Opening: context-setting declarative or a concept definition. Never opens with "I" or a personal anecdote.
Conclusion: 1-3 paragraphs, no "In conclusion" / "To sum up" markers.
Jargon: concept before name; academic terms mapped to code the reader knows.
Title: a thesis statement that conveys the argument.
Signature phrases (use naturally, don't force)
"easier to reason about", "source of truth", "data in, data out", "in the context of", "straightforward", "as opposed to".
Notes
Post drafts and ideas live under .ai/para/resources/blog-ideas/.
This is the personal jaysoo.ca voice. It is distinct from the Nx company blog — for that, see .ai/para/resources/architectures/nx-blog-architecture.md.
After invoking this skill, bump USAGE.md in dot-ai-config.
