Hashnode
Platform · this deep dive is built from the official pages of the platform itself and of the tools that support it.
Overview
Hashnode is a blogging platform for developers: you set up a "publication", write technical pieces and get a shared community feed for them. The platform states its purpose as "Write to think. Publish to connect" and describes its audience as "a community of builders, engineers and technical leaders who blog to sharpen their ideas".
What sets it apart from ordinary blog hosting in practice is headless mode and an open GraphQL API: you can use the platform as a store for your texts and build the front end yourself. What sets it apart from Dev.to and Medium is a custom domain and full programmatic access — but as of 2026 both have moved to the paid plan.
Worth knowing for planning: the platform was substantially rebuilt in 2026. Judging by the changelog, seventeen entries came out between February and July 2026 — from a redesign to putting the API behind a paywall. Articles about Hashnode from two years ago are not a reliable guide.
Who's here
The platform publishes no audience figures: neither the home page, nor the changelog, nor the terms of use gave us a single number for traffic, blogs or authors. The only quantitative statement about the community is indirect: tag pages now show the number of tag followers alongside the number of publications, which means the data exists but the platform does not disclose aggregate values.
The language is English: the interface and every document we checked are in English, and there is no language switcher on the open pages.
Getting started
Registration is standard, through the sign-in form on the home page. No document verification is required, and no separate "business account" is described on the open pages.
The age threshold is stated in the terms: "You must be at least 13 years of age to use the Service".
The key limitation of free entry: free accounts are capped at one publication, whereas the paid plan allows up to 10 publications (blogs). If you are planning several themed blogs, the only way through is to pay.
Moving an existing blog over is offered as bulk markdown import — also a paid-plan feature.
What you can publish
The format is Markdown articles. The platform discloses its technical limits not on the site but in its own open agent skill: no more than 15 tags per publication, a request body of up to 100 KB, image uploads of up to 8 MB, image/* only and no SVG. Formally these are API limits, but they also describe the maximum size of a piece.
What is prohibited is listed in the code of conduct: harassment and threats; hate speech on the grounds of race, gender, orientation, disability, religion or age; automated and bulk publishing, engagement farming, misleading links, SEO abuse; plagiarism and republishing other people's work without permission; malicious links and disclosure of other people's personal data.
On AI the rule is stated outright: "AI-generated content is permitted, but you must review and take responsibility for everything you publish". So text written by a model is not banned, but responsibility for it rests with you.
Rights to content stay with the author: "You retain full ownership of all content you create, publish, or share through the Service"; the platform is granted a non-exclusive, royalty-free, worldwide licence to host, display, distribute and reproduce it. When an account is deleted, the content is removed from public view.
How to grow
The feed is not chronological, and the platform admits it: the main feed uses a ranking algorithm instead of pure chronology, and recent posts with engagement rise higher. One specific signal is named in the same place: posts published on a custom domain get "a small boost", because, in the platform's wording, they usually turn out to be of higher quality. The practical conclusion follows: the paid plan affects not just features but visibility.
The platform lists its technical promotion layer separately: custom Open Graph images for tag and profile pages, schema.org JSON-LD markup on every public page, canonical tags and absolute addresses for OG images, and a web app manifest.
The platform has no ads manager: none of the pages we checked describes buying impressions or promoting posts for money.
Path to monetization
The platform does not pay authors. Neither the plan page, nor the terms of use, nor the changelog describes any share of ad revenue, any paid subscription to a blog, or any donations. The money flows the other way here: the author pays.
The price is stated plainly: $5 a month or $50 a year ("2 months free"). That buys you a custom domain, access to the GraphQL API, headless mode, the AI agent skill, AI editor tools, webhooks, backup to GitHub, bulk markdown import, hiding Hashnode branding, scheduled publishing, protection of posts from automatic removal, and up to 10 publications.
Note one discrepancy worth checking before you pay: the plan page names a flat $5 a month, while the launch announcement describes the price differently — "seat-based and adjusts automatically as members join or leave", that is, per team seat. For a solo author these are the same thing; for a five-person editorial team they are not.
There is good news for anyone who used the platform earlier: users who had a custom domain, webhooks or GitHub backup enabled before the paid plan launched keep those features for free.
Tools and automation
The official interface is GraphQL. The platform names the endpoint address in its own open repository: https://gql-beta.hashnode.com. Authorization is by personal token: the token is created in the dashboard (Account Settings → Developer / API tokens) and passed in the Authorization: Bearer header.
What you can do through it: publish and update posts, manage drafts, upload images via a two-step scheme with a presigned S3 URL, and read public data — posts, feeds, users, tags and publications.
The access boundaries changed in May 2026: reading used to be free, and now queries against your own publication require Pro. The platform names the reason outright — abuse by content scrapers and spammers. The agent skill clarifies the split: public reads (posts, feeds, users, tags) are free and require no token, while mutations and publication-scoped queries require an active Pro.
The platform publishes numeric API limits: maximum query depth is 10, the page size for most connections is 100 and 50 for drafts; the stated rate guidance is 20,000 requests per minute and 500 mutations per minute, though server-side rate limiting is not currently enforced; responses are cached for roughly up to 25 seconds.
There is a separate tool for machine reading: any article can be fetched as source Markdown — by appending .md to the address or by sending an Accept: text/markdown header; nothing changes for the browser.
For AI agents the platform has released an official skill, installed with the command npx skills add Hashnode/gql-skill; the source code is open.
Limits and rules
- Age. 13 and over.
- Free plan. One publication, no custom domain and no API.
- API. Mutations and publication-scoped queries only with an active Pro; the error comes back as
FORBIDDEN, and retrying the request is pointless. - Sizes. 15 tags, a 100 KB request body, an 8 MB image, no SVG.
- Automated moderation. It exists: the paid plan specifically promises that off-topic posts are exempted from automatic removal. Which means such removal is possible on the free plan.
- Content. Harassment, hate speech, bulk and automated publishing, engagement farming, SEO abuse and plagiarism are prohibited.
Who it's for
It suits engineers and technical teams who need a blog with decent technical SEO and the option of keeping their texts in version control: two-way GitHub sync and markdown import make Hashnode a sensible "warehouse" for articles.
It suits people building their own documentation site or front end: headless mode and GraphQL let you keep editing on the platform while serving the output yourself.
It does not suit you if you expect to earn money on the platform itself: there is no author payout mechanism here at all. It does not suit you if your budget is zero but you need a custom domain and automation: since May 2026 they cost money. And it does not suit non-technical subjects — the platform describes its own feed and community as engineering ones.
Verified data
The checked data this deep dive rests on.
About the platform
the platform describes itself like this: «Start a blog on your own domain, keep your audience, and reach readers across search and AI engines. Free to start.»
Self-description: this is the platform's claim about itself, not an independent assessment
source, checked 2026-07-28
Languages
the interface language of the home page: en
The language is taken from the markup attribute, the versions from hreflang
source, checked 2026-07-28
Catalogue section: all similar See also: catalogue index · region: USA and Canada · scheduler comparison · find by situation · platform restrictions