ShipSprint + Claude Code
One command in the terminal, and Claude Code can read your sprint board and update it, the same connection Claude and ChatGPT use, just from where you're coding.
The board, without leaving the terminal
Checking a ticket usually means alt-tabbing out of the editor, opening a browser, finding the right board, and reading a description that may already be stale by the time you get to it. ShipSprint speaks MCP, the open standard assistants and coding agents use to call real tools, and Claude Code connects to it the same way any MCP client does: one command in the terminal, and it has the same sixteen tools available inside a coding session that Claude and ChatGPT get from the web.
The practical use case is narrower and more concrete than a general chat assistant's: a developer mid-task in Claude Code can ask what a ticket actually says, log the hours they've spent on it, or leave a comment for whoever reviews the pull request, all without breaking flow to open another tab. The GitHub sync already moves and closes cards from branch and merge activity; this fills the gap in between, while the work is still in progress.
Real actions, from inside a coding session
The same tool calls the web assistants use, called from the terminal instead.
Claude Code calls update_issue's companion read tools to pull the ticket's real description and any comments, instead of working from a half-remembered summary.
Calls log_work and comment_on_issue directly. Both actions land in the workspace and the audit log, attributed to you.
Calls create_issue so the follow-up exists on the board immediately, rather than living in a comment or a mental note that gets lost.
Calls search_pages and read_page, pulling prior decisions straight into the coding session instead of a separate browser search.
Setup, permissions, and what's logged
Adding the server is a single command from your terminal, using the connection details in your ShipSprint profile, no browser-based connector flow required the way Claude.ai or ChatGPT use. Once added, it acts with your exact account permissions: nothing it can read or write inside the workspace goes beyond what you could do yourself in the app. Every action it takes, a comment, a logged hour, a new task, lands in the audit log exactly like any other, attributed to your account.
Full toolset: my_work, search, create_issue, update_issue, comment_on_issue, log_work, list_projects, list_issues, project_roadmap, whoami, list_spaces, list_pages, read_page, search_pages, create_page, update_page.
Where this genuinely isn't enough
This is a terminal-based MCP connection, not a visual board. If what you actually want is to look at the sprint board's layout, drag a card, or check a chart, that's still the web app's job; Claude Code can read and act on the same data, but it isn't a substitute for the visual view when that's genuinely what you need. It also carries your exact permissions, so it can't see or touch anything you personally don't have access to.
Common questions
Team and Business. It isn't included on the Free plan. See the pricing page for the full breakdown.
Same MCP server and the same sixteen tools, added with one terminal command instead of the one-click connector flow claude.ai uses. See the Claude integration page.
No, they do different jobs. GitHub sync moves and closes cards automatically from branch and merge activity. This MCP connection lets you read tickets and log actions mid-session, before that merge happens.
No. It acts with your exact account permissions, so anything outside your access in the app is outside its access here too.
Yes. The same actions the MCP tools expose are available as a plain REST API using personal access tokens, for teams that want to build their own tooling on top.
Related pages
See it on your own work.
A workspace your whole company will actually use is 60 seconds away. No card, no risk, nothing to install.
14-day full-access trial · sample project included · no card required