skills.rest
HugoRCD
HugoRCDCOMMUNITY
Explore all skills by this author →
Website𝕏@hugorcdGitHub

write-evlog-content

Reviews and rewrites evlog documentation, blog, landing, and skill prose against voice rules and AI-tell detection.

1.8k|61|Updated Jan 24, 2026
Search Tags:#linting#style guide#Technical Writing#Documentation#AI detection#evlog#Content Review
One-click install
npx skills add https://github.com/HugoRCD/evlog --skill write-evlog-content
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: write-evlog-content
Source: https://github.com/HugoRCD/evlog/tree/main/.agents/skills/write-evlog-content
Command: npx skills add https://github.com/HugoRCD/evlog --skill write-evlog-content

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Writing and reviewing prose for the evlog project requires consistent voice, accurate API references, verified competitor comparisons, and detection of AI-generated writing patterns. This Skill provides the complete rule set, terminology table, competitor dossiers, and scanner workflow so reviews produce cited findings instead of subjective taste.

Core Features & Use Cases

  • Structured content review: Runs the content-lint scanner, separates house-rule violations from rhythm judgments, and outputs findings with rule or tell IDs, verbatim excerpts, and a verdict (pass, minor, significant, blocked).
  • Constrained rewriting: Applies only the changes a finding names, preserves MDC structure and links, verifies code samples against packages/evlog/src, and returns passing pages unchanged.
  • Surface-specific rules: Carries dedicated rule files for docs pages, blog posts, the landing page, READMEs, changesets, SKILL.md files, and AGENTS.md files.
  • Use Case: Before publishing a new docs page on sampling, run the scanner on the file, review flagged AI tells against their legitimate twins, verify every backticked symbol exists in the source, and produce a verdict with cited findings.

Quick Start

Ask the agent to review a specific docs page such as apps/docs/content/2.learn/2.wide-events.md for voice, accuracy, and AI-generated tells, and return a verdict with cited findings.

Frequently Asked Questions about write-evlog-content

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I review evlog documentation for AI-generated writing?▼

Run pnpm content:lint on the target file to get metrics and findings, then judge each rhythm candidate against its legitimate twin in ai-tells.md. Every finding must cite a rule or tell ID, a verbatim excerpt, and what it costs the reader.

How do I check if a docs page matches the evlog source code?▼

Verify every backticked symbol and import path against packages/evlog/src and package.json exports. Phantom APIs like evlog/shared or evlog/browser are T-15 findings and always critical.

What surfaces does this content review cover?▼

It covers docs pages, blog posts, the landing page, package READMEs, changesets, SKILL.md files, and AGENTS.md files. Each surface has its own rule file loaded alongside the universal rules.

Can the content linter fix findings automatically?▼

Yes, pnpm content:lint --fix applies rules whose correction follows mechanically, such as retired entry points and redirected links. It never touches punctuation, and it reverts any file that scores worse after the fix.

When is a content finding critical versus standard?▼

Critical findings block publishing: wrong code samples, phantom APIs, claims the source contradicts, or landing promises no page keeps. Standard findings cover voice, rhythm, structure, and punctuation, and rhythm alone is never critical.

What are the limitations of automated content scanning?▼

The scanner cannot judge structure, verify competitor claims, or confirm that code samples run. It also excludes the skill's own references and Evi's operating skills, so those files require manual reading.

Related Skills

Explore More Content & Communication Skills→
home-assistanthome-assistant
9.7k

create-blog-post

Convert Google Docs Markdown drafts into Jekyll-ready Home Assistant blog posts.

Official
Intermediate
mia-cxmia-cx

setup-changesets

Configure and run changesets to publish versioned packages and changelogs.

Official
Intermediate
Be-CivicBe-Civic

bc-dossier-compilation

Generate a structured printable application dossier from completed Belgian administrative procedures.

Official
Advanced
giantswarmgiantswarm
7

improve-style

Apply style fixes and run Vale prose linting on markdown files.

Official
Intermediate