just say what you want.
the plugin gives your agent the same TODO list you see in the dashboard. your agent shows each item, offers what you can do with it, and asks. it never guesses.
- “what's on my plate?” a check-in across every blog: topics to schedule, keyword cards, research suggestions. then it asks what you want to do with each.
- “fill my week.” a plan first (topic a on day 1, topic b on day 2…). one yes from you, then it schedules them in order.
- “add these keywords, plus "habit tracker for adhd".” take all the suggestions, some of them, or your own. new topics show up on your next check.
- “schedule a research task for friday.” research takes up a day slot and costs a credit. your agent says so before it books it.
the interactive bits of bloggity. not the typing.
- 1
see the queue
your account, your blogs, your TODO list and your research tasks. same order as the dashboard.
- 2
schedule topics
put a topic on a day you pick, or the next open one. snooze it, park it in the backlog, bring it back.
- 3
feed the keyword list
accept suggested keywords, add your own, or "ask me tomorrow" when you're not in the mood.
- 4
steer research
accept, schedule, cancel, dismiss or backlog the research toady proposes.
it can't write or edit post bodies, change your voice, style, domain or billing, or show the published calendar. for those, the dashboard is one click away.
connect bloggity.
bloggity is a standard remote mcp server, so it works in any client that supports them. pick yours. you sign in with your bloggity account in the browser — no api keys, no passwords handed to your agent.
one command. then run /mcp inside claude code to sign in.
claude mcp add --transport http bloggity https://mcp.bloggity.io/mcp one command, then sign in when it asks.
codex mcp add bloggity --url https://mcp.bloggity.io/mcp add this to ~/.cursor/mcp.json (or .cursor/mcp.json in a project).
{
"mcpServers": {
"bloggity": {
"url": "https://mcp.bloggity.io/mcp"
}
}
} add a custom connector and paste the server url. sign in through the browser window that opens.
https://mcp.bloggity.io/mcp any client that supports remote (http) mcp servers with oauth works. use its "add remote server" option, or this config shape:
{
"mcpServers": {
"bloggity": {
"type": "http",
"url": "https://mcp.bloggity.io/mcp"
}
}
}
then ask your agent to run whoami. it triggers sign-in and confirms the account.
no bloggity account yet? the sign-in page has a “sign up here” button.
teach your agent your queue.
the bloggity skill is how your agent learns the routine: walk through each item, offer what you can do with it, check before it schedules anything, and keep an eye on your credits. connect, add the skill, done.
easiest way: let your agent set up both steps.
paste this into claude code, cursor, codex, grok or any agent that can edit its own config. it shows you every change and waits for your ok before touching anything.
prefer doing it by hand? copy
skills/bloggity
into your agent's skills folder (for example .claude/skills/). plugin installs bundle it for you.
Set up the Bloggity MCP server for me.
1. Add a remote (HTTP) MCP server named "bloggity" with URL
https://mcp.bloggity.io/mcp using whatever method this client supports
(its MCP config file or its "mcp add" command). It uses OAuth, so do not
ask me for an API key or token.
2. Tell me the exact change you are about to make and which file or setting it
touches, and wait for my OK before writing anything.
3. After it is added, tell me if I need to restart the client or sign in via
my browser, then call the "whoami" tool to confirm the connected account.
4. Install the Bloggity skill. Fetch
https://github.com/vumpt/bloggity-mcp/blob/main/skills/bloggity/SKILL.md,
show it to me, then save it in this client's skills/instructions location
(for example .claude/skills/bloggity/SKILL.md or .agents/skills/bloggity/SKILL.md).
If this client has no skills support, add its contents to my agent
instructions file instead.
Only use the URL above. Do not run downloaded scripts or install anything else. your account, your say.
scoped to your team
every call only touches your own team's blogs. read access lists; write access changes things, and your client asks for it separately.
credits stay yours
scheduling is never blocked by credits, and generating a post early costs one. your agent asks first, every time.
revoke any time
cut off any connected client under connected apps. it takes effect immediately.
toady's ready when your agent is.
connect once, then manage every blog you run from wherever you already work.