dev-repurpose
Turn one long asset (blog post, docs, newsletter, transcript, client call) into a dev.to article with canonical URL back to the source. Produces title options, tags, body markdown, cover prompt, and a copy-ready block. Never publishes without an explicit yes.
naheel0 · 已有静态检查记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add naheel0/dev-skills --skill dev-repurpose技能说明
name: dev-repurpose description: >- Turn one long asset (blog post, docs, newsletter, transcript, client call) into a dev.to article with canonical URL back to the source. Produces title options, tags, body markdown, cover prompt, and a copy-ready block. Never publishes without an explicit yes.
dev-repurpose
Treats dev.to as a syndication channel, not the source of truth. The user
writes once (on their own blog / docs / newsletter) and repurposes to dev.to
with canonical_url pointing to the original — preserving SEO and making
dev.to a safe distribution surface.
Before you write
- Read
~/.claude/devto/voice.mdif it exists. - Confirm the user owns the source asset (or has rights to share it). dev.to's plagiarism policy prohibits all plagiarism types — direct, self, mosaic, accidental. Users certify rights to all posted content.
- Ask for the source URL (the canonical home).
- Ask whether the original has already been indexed / published. Best
practice: let the original index a few weeks first, then syndicate to
dev.to with
canonical_urlback to it.
The shape
Same as dev-post, plus canonical handling:
Title <10 words, keyword + number. Can differ from source title.
Body Adapted for dev.to: H2/H3, short paragraphs, code blocks,
cover image, CTA/question at end. Not a verbatim copy — rewrite
the intro for dev.to's audience.
Cover 1000×420 absolute URL (new cover, not source's).
Tags exactly 4, mixed broad + niche.
Canonical canonical_url = source URL. Set on dev.to copy, not source.
AI disclosure| if AI assisted, add #ABotWroteThis or in-article statement.
Cross-posting flow (recommended)
- Publish on own blog (canonical home). Let it index 1–2 weeks.
- Repurpose to dev.to with
canonical_url→ own blog. - Optionally repurpose to Hashnode (canonical URL) and Medium (import from URL sets canonical automatically).
- Always set
canonical_urlto the primary blog to consolidate SEO. - Add an explicit link back to the original in the dev.to body text.
The loop
1. Read the source asset. Extract the core idea, code snippets, and
takeaways.
2. Pick three titles. Run through title-formulas.json (from dev-post).
3. Draft the dev.to adaptation. Not a copy — rewrite for dev.to's
beginner–intermediate audience.
4. Humanize. Run through /dev-human.
5. Tags. Exactly 4 from tag-suggestions.json.
6. Print the block. Copy-ready, with canonical_url in front matter or
gear icon. Then:
POST READY
title: #12 Comparison · Adapted title
length: 1,100 words · ~4 min read
humanizer: 5 artefacts stripped, human score 84 PASS
tags: javascript, webdev, tutorial, react
cover: {{cover image URL}}
canonical: https://yourblog.com/original-post
series: {{none}}
published: draft
post at: Tuesday 9:00am ET
Reply "yes" to log it, or tell me what to change.
7. Never publish automatically. Copy-paste into dev.to editor, OR use
Forem API POST /api/articles if DEVTO_API_KEY is present (deferred feature).
Record canonical and series in ~/.claude/devto/log.md so /dev-learn
can track whether syndicated posts (canonical set) underperform native ones.
Rules
- Canonical on dev.to copy, not source. Don't set canonical to dev.to on the source blog (SEO dilution).
- Substantial content required. dev.to's Terms §11: posts must contain substantial content — may not merely reference an external link holding the full post.
- Affiliate links disclosed. Clearly state compensation relationships.
- No promotional backlinks. Main purpose can't be SEO backlink farming (except personal blog or company blog posted under that company's org).
- No AI-generated educational content the human didn't already understand.
- Never fabricate metrics or outcomes.
Skipped (ponytail)
- No RSS import automation (dev.to has built-in RSS import; drafts only, manual publish).
- No auto-sync of upstream edits (RSS import doesn't auto-sync).