Hashnode has shipped the smallest possible setup for an agent to run a blog. In Hashnode MCP server: manage your blog from Claude, ChatGPT, and Cursor, the post describes a remote server that lets AI clients “publish, draft, edit, search, and delete posts” on a Hashnode blog. “Add one URL, log in with your Hashnode account, and you’re connected. No API token, nothing to install.” In Claude Code it is one command; in ChatGPT and other clients, the same URL pasted where the client asks for a remote server, after which “your browser opens Hashnode. Log in and approve access.”
There are no figures, and the access model is the substance. “You never handle a token. Every client you connect shows up in Developer Settings, and you can disconnect any of them at any time. Access stops immediately.” The tool list follows the publication’s existing roles: contributors “can’t publish directly,” so an agent submits a draft for review and an owner or admin publishes it, with tools for submitting, rejecting and listing drafts awaiting review. The post also gives the simplest safety advice in the series: “Posts go live immediately. To look first, ask your agent to save a draft, then publish it when you’re ready.” And it draws the line between surfaces cleanly: the full GraphQL API through a skill for coding agents, the MCP server for chat clients or anyone who would rather not manage a token.
The catalog holds the API underneath. The Hashnode provider page carries the Hashnode GraphQL API. On the Agent Readiness score, spec presence, auth clarity, error semantics, the rate limit signal, the event surface and agent skills are lit. The MCP server is unlit, as are delegated identity, protected resource metadata and dynamic client registration, which is what a log-in-and-approve remote server without a token typically implements. The launch is newer than the record’s last probe.
The Kin Score is 56.9, strong band, with developer ergonomics at 78.6 and operational transparency at 63.2 against contract governance at 4.5. The Agent Readiness score is 28.8, agent-ready, and it should move on the next pass if the server publishes the metadata its login flow implies. Two details deserve credit beyond the launch. A draft-first workflow is a dry run expressed in the product, and the reviewer roles are a human in the loop built into the publication rather than bolted onto the agent. Hashnode has given agents a way in that never puts a secret in a transcript. The record needs one more probe to see it.