devto — Claude Code skill: what it does, how to install
Publishes articles to dev.to, edits existing posts, and pulls view and reaction stats using the Forem API v1.
Apache-2.0 213 installs ★ 17 updated 2026-10-25
npx skills add AceDataCloud/Skills --skill devto -a claude-codeWhat devto does
This skill interacts with dev.to via Forem API v1 endpoints using curl and jq. The agent checks credentials against https://dev.to/api/users/me, submits Markdown articles as drafts or live posts, and modifies existing entries using PUT requests.
Every call relies on the $DEVTO_API_KEY environment variable. Without this token exported in your environment, all requests fail with a 401 status.
Cross-posting workflows can include a canonical_url to keep search ranking on the primary source. The platform caps tags at four items in lowercase without spaces, and firing updates too quickly triggers a 429 rate-limit error.
What it helps with
- post Markdown files to dev.to as drafts or live entries
- check view counts, comment totals, and reaction metrics for published posts
- update published articles using their dev.to post ID
- preserve SEO attribution by passing a canonical URL during cross-posting
Example prompts for the agent
- Publish article.md to dev.to as a draft with tags webdev and ai.
- Show view counts and reactions for my recent dev.to posts.
- Update dev.to post 12345 with the revised content in article.md and set it live.
How to install devto
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill devto -a claude-code2Manually: put the files in place
~/.claude/skills/devto/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/devto/
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill devto -a codex2Manually: put the files in place
.agents/skills/devto/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill devto -a cursor2Manually: put the files in place
.cursor/skills/devto/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install AceDataCloud/Skills skills/devto2Manually: put the files in place
.github/skills/devto/SKILL.md1Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill devto -a gemini-cli2Platform CLI
gemini skills install https://github.com/AceDataCloud/Skills --path skills/devtoManually: put the files in place
.devin/skills/devto/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add AceDataCloud/Skills --skill devto -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about devto
How to install devto in Claude Code?
Run npx skills add AceDataCloud/Skills --skill devto -a claude-code.
What does devto do?
It manages your dev.to publishing workflow through Forem API v1. You can post Markdown drafts, review performance metrics, and edit articles from your terminal.
What are the tag constraints when publishing?
You can add up to 4 tags per article. Every tag must be lowercase and contain no spaces.
Source text
SKILL.md · Apache-2.0
---
name: devto
description: Publish, update and read articles on DEV (dev.to) via the Forem API v1. Use when the user mentions dev.to / DEV Community, publishing a blog post to dev.to, cross-posting an article, updating a published post, or listing their dev.to articles and stats.
when_to_use: |
Trigger when the user wants to publish a Markdown article to their
dev.to account, update a previously published article, or list /
inspect their own dev.to articles (views, reactions, comments). The
connector stores a DEV API Key with full account access — confirm
before publishing publicly (you can publish as a draft with
published=false first).
connections: [devto]
allowed_tools: [Bash]
license: Apache-2.0
metadata:
author: acedatacloud
version: "1.0"
---
Call the **Forem API v1** (dev.to) with `curl + jq`. The user's API key is in
`$DEVTO_API_KEY`; every call needs the headers `api-key: $DEVTO_API_KEY` and
`Accept: application/vnd.forem.api-v1+json`. Base URL: `https://dev.to/api`.
Errors come back as JSON with an `error` / `status` field — show them verbatim.
`401` means the API key is invalid → the user must re-connect the DEV connector.
**Always start by confirming the key** and learning the account:
```bash
curl -sS -H "api-key: $DEVTO_API_KEY" -H "Accept: application/vnd.forem.api-v1+json" \
"https://dev.to/api/users/me" | jq '{username, name}'
```
## Publish an article
**Confirm with the user before publishing publicly.** Default to a draft
(`published:false`) unless they explicitly say publish/now.
```bash
TITLE="My title"
BODY_MD="$(cat article.md)" # full Markdown body
jq -n --arg t "$TITLE" --arg b "$BODY_MD" \
'{article:{title:$t, body_markdown:$b, published:false, tags:["ai","webdev"]}}' \
| curl -sS -X POST "https://dev.to/api/articles" \
-H "api-key: $DEVTO_API_KEY" \
-H "Accept: application/vnd.forem.api-v1+json" \
-H "Content-Type: application/json" \
-d @- \
| jq '{id, url, published}'
```
To publish a draft later (or edit), `PUT /api/articles/{id}` with the same
shape (set `published:true`). Front-matter inside `body_markdown` (a `---`
block) can also carry `title`, `tags`, `series`, `canonical_url`, `cover_image`.
- **Canonical URL:** when cross-posting, set
`"canonical_url":"https://your-blog/original"` so DEV points SEO back to the
source — important for the article→video / cross-publishing flow.
## List / inspect my articles
```bash
# My published + draft articles (paginated; per_page max 1000).
curl -sS -H "api-key: $DEVTO_API_KEY" -H "Accept: application/vnd.forem.api-v1+json" \
"https://dev.to/api/articles/me/all?per_page=30" \
| jq '.[] | {id, title, published, page_views_count, public_reactions_count, comments_count}'
# A single article's full content + stats.
curl -sS -H "api-key: $DEVTO_API_KEY" -H "Accept: application/vnd.forem.api-v1+json" \
"https://dev.to/api/articles/ARTICLE_ID" | jq '{title, url, reactions: .public_reactions_count, views: .page_views_count}'
```
## Gotchas
- **Tags:** max 4, lowercase, no spaces (e.g. `webdev`, `machinelearning`).
- **Rate limit:** article create/update is throttled (a few per 30s); space out
bulk publishes or you'll get `429`.
- `body_markdown` is the source of truth — if you put a `---` front-matter block
at the top, its fields override the JSON `article` fields.
## Record the output
After you successfully publish and obtain the live result URL, call the built-in
`publish_artifact` tool ONCE so the user can track this deliverable in **My Outputs**:
```
publish_artifact(kind="article", channel="devto", title="<title>", url="<the REAL returned URL>", status="delivered")
```
Use the real returned URL — never fabricate one. Call it once per published item,
only after delivery is confirmed; skip it (or use `status="failed"`) if publishing failed.
See `_shared/artifacts.md`.
© acedatacloud · Apache-2.0 · AceDataCloud/Skills