TL;DR: You can still auto publish to Hashnode through its official GraphQL API, but since May 13, 2026, every request (reading or writing) requires a Hashnode Pro subscription on the target publication. Pro launched June 11, 2026, at $5 a month or $50 a year, and bundles the API in with custom domains, scheduled publishing, and webhooks. If you don’t want to pay per publication just to automate posting, driving the same logged-in browser session you already write in gets you the same result.
Open Hashnode’s API docs today and you’ll still find the same two-mutation publishing flow developers have used for years: createDraft, then publishDraft. What changed is the gate in front of it. Run either mutation from a free account now and the response is an authorization error, not a published post. Hashnode retired free GraphQL access on May 13, 2026, citing scrapers and spammers that were mirroring content to spam farms at scale through the same endpoint developers used for legitimate automation. If you’re running a script that posted fine in April, it stopped working in May, and no amount of debugging your token fixes that.
What changed on Hashnode’s side
Two announcements landed a month apart, and together they explain why “auto publish to Hashnode” now has a price tag attached:
- May 13, 2026: Hashnode’s own changelog states it plainly: “We’re retiring free GraphQL API access. Every API request, queries and mutations, now requires a Pro plan on your publication.” Write access already needed a paid plan before this; the change was making read-only queries paid too, closing the loophole scrapers were using.
- June 11, 2026: Hashnode Pro launched at $5 a month or $50 a year, billed per publication and seat-based for teams. The GraphQL API is one line item in a bundle that also includes custom domain mapping, a headless CMS mode, an AI writing assistant, webhooks, GitHub backup, bulk Markdown import, branding removal, and scheduled publishing. Free accounts keep the web editor and one publication; Pro raises that cap to ten.
If you already had webhooks, a custom domain, or GitHub backup switched on before Pro existed, Hashnode grandfathered those in at no extra cost. Everyone starting fresh after June 11 pays for the whole bundle to get any one piece of it, the API included.
What the API still gives you if you pay for it
Assuming your publication is on Pro, the mechanics haven’t changed much from what developers have documented for a couple of years. createDraft takes a title, Markdown body, and publication ID and hands back a draft ID; publishDraft takes that draft ID and returns the live URL. You can’t skip straight to a published post in one call.
A few details trip people up on the first attempt:
- Tags must be objects, not strings. Hashnode expects
{ name, slug }pairs with lowercase, hyphenated slugs. Send a plain array of tag names and, as one developer documented after debugging it, the API silently accepts the request without applying any tags, no error thrown. - The publication ID isn’t your username. It’s a separate internal identifier you pull from the dashboard or a preliminary query, and every mutation needs it.
- Cover images go through
coverImageOptions.coverImageURL. The API wants a URL to an already-hosted image, not a file upload, so you still need somewhere to host the image before you call the mutation. - Canonical URLs use
originalArticleURL. Set it when you’re syndicating a post that lives somewhere else first, so Hashnode doesn’t compete with your own site for the same content in search results.
None of that is exotic if you’re already comfortable with GraphQL. What changed is that testing any of it now requires a $5-a-month subscription just to see whether your request went through.
Let an AI run it through your own account instead
If you’re posting to Hashnode as one stop in a five- or six-platform rotation, paying for API access on each platform that gates one adds up fast, and most of them gate different things: Medium closed its API to new integrations entirely, dev.to’s stays free but caps you at 10 requests per 30 seconds, and now Hashnode charges per publication. PublishPort skips that per-platform negotiation. It drives the same browser session you already use to write in Hashnode’s own editor, the one that’s free on every plan, so an AI agent gets a local_bash command and a one-line capability description instead of a hand-rolled API client that needs its own paid tier.
That session-based model is the same one covered for automating dev.to posts, just pointed at a platform where the official API now has a subscription attached. Download PublishPort and the AI publishes through your logged-in account, cover image, tags, and canonical link included, without your publication needing to be on Hashnode Pro just to automate posting.
Setting it up
- Install the PublishPort client and let it use the Chrome profile you’re already logged into.
- Confirm Hashnode shows as connected in the client once it detects your existing session; there’s no separate app registration or token to generate.
- Give the AI a draft and let it handle formatting: title, Markdown body, tags, and a cover image URL, matching what Hashnode’s own editor expects.
- Let it publish through the same flow you’d click through by hand, then move to the next platform in your rotation without a second login.
- Check the live post the same way you’d check any manual publish, since PublishPort runs through your real account rather than a shared server.
Limits and what this doesn’t do
Be clear-eyed about what a browser session covers and what it doesn’t. Hashnode’s own scheduled publishing, its AI writing tools, custom domains, and webhooks are still Pro features gated at the platform level; driving the editor doesn’t unlock those, because they aren’t part of the editor’s manual publish flow to begin with. PublishPort’s Hashnode support also doesn’t pull analytics back the way it does for Medium or dev.to, since Hashnode’s own dashboard doesn’t expose that through the same interface. And this is publishing through your real account at whatever pace a person would use it, not bulk-posting or duplicate blasting across publications. No automation avoids Hashnode’s own content and spam rules, and nothing here promises your account is immune to platform enforcement.
Before you automate Hashnode posting, check:
- Is your publication actually on Pro, or were you assuming the old free API still works?
- Do your tags arrive as
{name, slug}objects, not a plain string array? - Is your cover image hosted somewhere with a stable URL, since the API won’t accept a raw upload?
- Are you setting
originalArticleURLwhen this post already lives on your own site first? - If you’re posting to five other platforms too, is a per-platform API subscription cheaper than one session-based tool?
FAQ
Is Hashnode’s API still free?
No. As of May 13, 2026, every GraphQL API request, including read-only queries, requires the target publication to be on Hashnode Pro. Before that date, reading was free and only writing needed a paid plan; now both do.
How much does Hashnode Pro cost?
$5 a month or $50 a year per publication, billed per seat for teams. It launched June 11, 2026, and bundles the GraphQL API with custom domains, scheduled publishing, webhooks, GitHub backup, an AI writing assistant, and up to ten publications per account.
Can I still schedule posts on Hashnode without paying?
Scheduled publishing moved into the same Pro bundle as the API. The free tier keeps the standard web editor and manual publish button, but future-dated scheduling is a Pro feature now.
Why does my Hashnode API request return an authorization error?
The most common cause since May 2026 is a publication that isn’t on Pro. Double-check your subscription status in Billing before assuming your token or mutation syntax is wrong.
Can I cross-post to Hashnode without hurting my site’s SEO?
Yes, set the originalArticleURL field to your original post’s URL when you publish or update through the API. Hashnode treats that as the canonical source, so search engines credit your site rather than treating the two as duplicate content.
Do I need Hashnode’s API to auto-publish there at all?
No. The API is one way in, but it’s not the only one. Tools that drive a logged-in browser session, PublishPort among them, publish through the same editor a person would use manually, which doesn’t require your publication to carry a separate Pro subscription just for automation.