Mastodon posts with Claude Code and Bolta
Turn your repo's release notes into social drafts, from the terminal. Parts of the Claude Code route are not confirmed yet, and the note below says which. Nothing goes live on Mastodon until you say yes.
Partly confirmed, see what is and is notChecked 2026-11-21
Two slash commands, then a browser sign-in. You tick the brands Claude Code may reach.
Read the top entry in CHANGELOG.md and draft a Mastodon post about it inside my instance's limit
Illustration of the reply: Bolta returns a draft showing the connected account and the scheduled time, with Approve, Edit and Reject actions.
What is and is not confirmed
Read this before you set up Claude Code
- Confirmed in Claude Code's official docs: remote HTTP MCP servers with browser sign-in through /mcp, skills as SKILL.md folders, and plugin marketplaces added from a GitHub repo.
- Bolta's own plugin is the most tested path. A full real-world run on every setup is still beta (plugin v0.5.3).
- Not confirmed here: the exact minimum Claude Code version for plugin marketplaces (the Bolta plugin README names 2.1.295 or later only as an example), and running Bolta end to end from claude -p in scripts. Claude Code's docs say headless runs cannot do the browser sign-in. The API key header route follows Claude Code's documented --header option but was not run end to end.
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.
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 Code: 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.
| What to know | Mastodon's own rule | What Bolta does |
|---|---|---|
| Post length | Not 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. |
| Visibility | Not stated on this page. | Public unless you set another visibility. |
| Formatting | Not stated on this page. | Mastodon posts are plain text here. Markdown is flattened, and #hashtags and @mentions stay as typed. |
| Finding conversations | Not stated on this page. | Hashtag-based. The query is folded into a single tag, because full-text status search is unavailable on most instances. |
Claude Code jobs for Mastodon, and how to write them
- Draft a week of build-in-public posts in your own voice between coding sessions
- Review what Bolta has queued without leaving the terminal
- Ask for your instance's limit before drafting. It differs from instance to instance.
Example prompts for Mastodon
Things to type in Claude Code. Each one asks for a draft and shows you the text first.
Read the top entry in CHANGELOG.md and draft a Mastodon post about it inside my instance's limit
What is waiting for my approval in Bolta right now?
Turn the release notes into a Mastodon post with one relevant hashtag and show it to me first
Which Bolta brands can you see, and which accounts are connected to each? Do not change anything.
Bolta saves a draft, you read the exact text, you say yes
- 01
Bolta saves a draft
Claude Code asks Bolta for the Mastodon post in your brand voice (save-post). A saved draft is never public.
- 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.
- 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.
Using Claude Code for Mastodon
Parts of this route are not confirmed. The note above says which.
For Mastodon, Claude Code, from Anthropic, reaches Bolta through Plugin from the bolta-skills marketplace, or claude mcp add. Sign-in: Sign in (OAuth) from /mcp. The Mastodon account itself is connected inside Bolta, not in Claude Code.
Claude Code is a terminal coding agent, so the natural Mastodon job is turning a changelog, README or release notes already in your repo into a draft. Bolta's plugin for it bundles the connection and a skill.
An illustration from the Claude Code guide, not real output. You ask: Which Bolta brands can you see, and which accounts are connected to each? Do not change anything. The reply goes along these lines: These are the brands you ticked when you signed in, with their connected networks listed under each. If a brand is missing, run /mcp, pick Bolta and sign in again to tick it. Read only. Nothing changed.
An illustration from the Claude Code guide, not real output. You ask: Send the three drafts you just saved to our reviewer for approval, and do not publish anything. The reply goes along these lines: I looked up who on the brand can receive approval requests and listed them for you. Tell me which person, and I will send the drafts to them for review. Sending for review does not publish. The reviewer decides. Waiting on you to pick a reviewer.
The plugin installs the connection and the skill. If you already added Bolta yourself, installing the plugin on top leaves two Bolta entries on one organization. In that case add only the skill with npx skills add boltaai/bolta-skills --skill bolta.
Prefer sign-in. Posts from a sign-in are held as AI drafts for review, and a pasted Bolta API key skips that hold.
One connection is one organization. Agencies tick their client brands in the same connection. Several organizations need one connection each.
- Draft from your repo. Changelog, README and release notes become drafts in your voice and the network's rules, saved without publishing.
- Write a batch of drafts. Ask for several posts at once, such as a week of build-in-public notes, and review them together.
- Find which brand to use. It lists the brands you ticked at sign-in. With more than one, it asks which you mean instead of picking for you.
- Check connected accounts. It reads the networks connected to that brand, so it will not draft for one you have not connected.
- Setup: Start Claude Code and find its prompt. Open a terminal and start Claude Code. The next two commands are typed into the Claude Code prompt, not into the terminal itself.
- Setup: Add the Bolta marketplace. Run /plugin marketplace add boltaai/bolta-skills. A marketplace is a list of plugins Claude Code can install, and boltaai/bolta-skills is the GitHub repository that holds it. Run /plugin and look for bolta-marketplace to check it worked.
- MCP with an API key header. Claude Code documents --header and .mcp.json headers, and Bolta accepts keys starting with bolta_sk_. We have not run it end to end. A pasted key skips the AI-draft hold.
- A routine you confirm and switch on runs in Bolta on its schedule, even with Claude Code closed.
- Sign in, then ask. Run /mcp, select Bolta and sign in in your browser. Back in the chat, type: set me up.
Connect Claude Code to Bolta
Two slash commands, then a browser sign-in. You tick the brands Claude Code may reach.
Mastodon questions, answered for Claude Code
- What is the Mastodon character limit for a post from Claude Code?
- 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 Code public?
- Yes, unless you set another visibility. Say who may see the post when you ask, and check it when you read the draft.
- Claude Code does not see the Bolta tools.
- Run /reload-plugins if the install summary asked for it, then run /mcp and confirm Bolta shows as connected. A plugin installed from the terminal with claude plugin install loads the next time you start Claude Code.
- I see two Bolta entries in /mcp.
- You installed the plugin and also added Bolta yourself. Keep one and remove the other. If you keep your own connection, add only the skill with npx skills add boltaai/bolta-skills --skill bolta.
- A brand I expect is missing.
- It was not ticked at sign-in, or you added it afterwards. Run /mcp, pick Bolta, sign in again and tick it. If you ticked several brands, name the one you mean in your request.
- Claude says the sign-in expired or you are not authenticated.
- Run /mcp, pick Bolta and sign in again. Sign-ins last 60 days and normally renew themselves. Claude should not retry in a loop.
- Does Claude Code publish Mastodon posts without my yes?
- No. Nothing is approved, scheduled or published until you say yes to that Mastodon post. Whether Claude Code completes the connection is not fully confirmed, so read the note above first.
More from Claude Code
Claude Code 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 Code
Connect Claude Code, ask for a Mastodon draft and read the exact text. Nothing goes live until you say yes.