Claude Cowork + Mastodon

Mastodon posts with Claude Cowork and Bolta

Run your weekly social media routine from Claude Cowork. Parts of the Claude Cowork route are not confirmed yet, and the note below says which. Nothing goes live on Mastodon until you say yes.

Add a custom connector in Customize, then sign in with your Bolta login. No API key, no terminal.

Claude CoworkBolta connected
You

Tell me the character limit for my Mastodon instance, then draft a post about our new release inside it

Illustration of the reply: Bolta returns a draft showing the connected account and the scheduled time, with Approve, Edit and Reject actions.

Illustration. Your drafts show your real text, account and time.

What is and is not confirmed

Read this before you set up Claude Cowork

  • Confirmed from Anthropic's help pages: custom remote MCP connectors work in Claude, Claude Desktop and Cowork, using one set of steps for all three. Bolta's server address and sign-in match that route.
  • Confirmed: Cowork needs a paid Claude plan (Pro, Max, Ultra, Team or Enterprise), and recurring tasks are started with /schedule.
  • Not confirmed by Bolta: the full path inside Cowork has not been tested end to end by us, and Anthropic gives no Cowork-only menus, so labels below may differ slightly in your version.
  • Not confirmed: whether a scheduled task can finish Bolta tool calls with nobody at the keyboard. Anthropic only says a tool set to Always ask can stall a run.
  • Bolta is not listed in Anthropic's connector directory. The custom connector is the only route described here.
Bolta on Mastodon

What Bolta does on Mastodon

The text limit on Mastodon is set by your own instance, and Bolta reads it for the connected account instead of using one fixed number. The usual limit is 500 characters.

  • Text posts, within the limit your instance sets.
  • Public posts by default, or another visibility if you set it.
Limits and format

Mastodon limits and format rules

Bolta's posting rules for your workspace (get-posting-rules) win if they differ from this page.

Not promised for Mastodon from Claude Cowork: Polls, threads and instance-specific features. None is confirmed for Bolta. Any rule from a particular instance. Each instance sets its own, so check yours.

Mastodon limits and format rules
What to knowMastodon's own ruleWhat Bolta does
Post lengthNot stated on this page.Set by your instance. Bolta reads it per account, and the usual limit is 500. Text over the limit is trimmed back to the last whole sentence or word that fits.
VisibilityNot stated on this page.Public unless you set another visibility.
FormattingNot stated on this page.Mastodon posts are plain text here. Markdown is flattened, and #hashtags and @mentions stay as typed.
Finding conversationsNot stated on this page.Hashtag-based. The query is folded into a single tag, because full-text status search is unavailable on most instances.
Best practices

Claude Cowork jobs for Mastodon, and how to write them

  • Point Cowork at a folder of briefs on your computer and turn each one into a draft post
  • Ask Cowork each Friday what is still waiting in your Bolta inbox
  • Ask for your instance's limit before drafting. It differs from instance to instance.
Prompts

Example prompts for Mastodon

Things to type in Claude Cowork. Each one asks for a draft and shows you the text first.

Tell me the character limit for my Mastodon instance, then draft a post about our new release inside it

Read the briefs in my Launch folder and save one draft post per brief

Draft a Mastodon post that is visible to everyone, and show me the text before saving it

Check my Bolta inbox and give me one line per item. Do not approve anything.

Review

Bolta saves a draft, you read the exact text, you say yes

  1. 01

    Bolta saves a draft

    Claude Cowork asks Bolta for the Mastodon post in your brand voice (save-post). A saved draft is never public.

  2. 02

    You read the exact text

    On Mastodon the exact text includes its visibility, because the default is public, and where the limit for your instance would cut it.

  3. 03

    You say yes

    Only then does Bolta schedule or publish (publish-post), in your workspace's timezone. Safe Mode, client sign-off and brand approval rules still apply.

Claude Cowork and Bolta

Using Claude Cowork for Mastodon

Parts of this route are not confirmed. The note above says which.

For Mastodon, Claude Cowork, from Anthropic, reaches Bolta through Custom connector (hosted MCP server). Sign-in: Sign in (OAuth). The Mastodon account itself is connected inside Bolta, not in Claude Cowork.

Cowork is Anthropic's desktop and cloud agent for paid Claude plans. Its /schedule command starts recurring tasks, so a weekly Mastodon drafting job can live there, with Bolta still holding every draft for your yes.

An illustration from the Claude Cowork guide, not real output. You ask: Every Monday at 9am draft five posts for Northwind and keep them in review. The reply goes along these lines: Here is the plan: weekly on Monday at 9:00 in your workspace's timezone, five drafts, left in review. It repeats even with this app closed. Say yes to save it. It stays switched off until you separately ask me to activate it. Plan shown. Waits for your yes.

Cowork is Anthropic's agent for knowledge work. You give it a task, it works through the steps, and it can use connectors, skills and plugins along the way. It needs a paid Claude plan.

Bolta's Claude setup guide says media goes in as a public link. If a post needs an image, give Cowork a public link to it. A file that only sits on your computer is not documented to attach.

Bolta's rule is that nothing is approved, scheduled or published until you say yes, and for a single post that means you see the exact text first. Cowork carries that rule, because the Bolta connector tells it to.

  • Review what is waiting. Ask what is pending and Cowork reads the Bolta inbox. Approving needs your yes in words, for the exact text, and a rejection goes in with your own reason.
  • Voice and brand. Cowork can read your saved voice and brand details, and can help you set them up from your website, with your confirmation.
  • Cowork finds your brands. The task starts by asking Bolta which organizations and brands the connection may use, so it never guesses where a post belongs.
  • It loads your voice and brand context. Next it reads the voice and brand details saved in Bolta, so the wording starts from how you write.
  • Setup: Open Connectors in Customize. Go to Customize, then Connectors, click Add and pick Add custom connector. On Team or Enterprise the Owner adds it instead, under Organization settings, then Connectors, and each member then clicks Connect.
  • Setup: Sign in with your Bolta account. Click Add, then sign in with your Bolta login in the browser window that opens. You do not need an API key or a terminal.
  • Anthropic's connector directory. Bolta is not listed there as far as we have confirmed, so the custom connector is the only route this page describes.
  • Anthropic says a connector tool set to Always ask may make a scheduled run wait for your answer or finish without that tool.
  • Add Bolta as a custom connector. Open Customize and go to Connectors. Click Add, choose Add custom connector, call it Bolta, paste https://mcp.bolta.ai/mcp, then sign in with your Bolta login.
Connect

Connect Claude Cowork to Bolta

Add a custom connector in Customize, then sign in with your Bolta login. No API key, no terminal.

FAQ

Mastodon questions, answered for Claude Cowork

What is the Mastodon character limit for a post from Claude Cowork?
It depends on your instance. Bolta reads the limit for the connected account, and the usual limit is 500 characters. Anything over it is trimmed back to the last whole sentence or word.
Are Mastodon posts from Claude Cowork public?
Yes, unless you set another visibility. Say who may see the post when you ask, and check it when you read the draft.
Does Claude Cowork work with Bolta?
It should, through a custom connector at https://mcp.bolta.ai/mcp. Anthropic documents custom remote connectors for Cowork, and Bolta's address and sign-in fit that route. Bolta has not tested every Cowork menu end to end, so this page marks that plainly.
Why did my scheduled task stop waiting for approval?
Anthropic says a scheduled task using a connector tool set to Always ask may wait for your answer or finish without that tool. Set trusted read and draft tools to Allow always, and leave approve, schedule and publish on the setting that asks.
Cowork says the server cannot be reached.
Check the address is exactly https://mcp.bolta.ai/mcp. Claude connects from Anthropic's cloud, so the server must be on public HTTPS port 443. A local or private address will not work.
My image did not attach to the draft.
Bolta's setup guide says media goes in as a public link. Give Cowork a public link to the image instead of a file on your computer.
Does Claude Cowork publish Mastodon posts without my yes?
No. Nothing is approved, scheduled or published until you say yes to that Mastodon post. Whether Claude Cowork completes the connection is not fully confirmed, so read the note above first.

More from Claude Cowork

Claude Cowork can work with the other networks Bolta publishes to. Each page has its own limits and review notes.

Mastodon from other AI agents

The same Mastodon limits apply whichever agent asks. The connection and the setup are what change.

Try a Mastodon draft from Claude Cowork

Connect Claude Cowork, ask for a Mastodon draft and read the exact text. Nothing goes live until you say yes.