On claude.ai, Agenticboard is a custom connector: you add it once, sign in, and approve what Claude may reach. There is no token to copy and nothing to keep on disk, because the connection authenticates over OAuth 2.1 and Claude receives a scoped token behind the scenes. From then on Claude reads and drives the same board, tickets and docs your team uses.
In claude.ai, open Settings → Connectors → Add custom connector. Name it — Agenticboard does nicely — and set the URL to https://agenticboard.app/api/mcp/. Leave the OAuth Client ID and Secret blank; the server registers the client for you.
Click Add, then Connect. You are sent to Agenticboard to sign in and approve the connection. On the consent screen you pick the same three things a token would carry: the access level, which projects Claude may reach, and the documentation scope. Then choose Authorize.
Ask Claude to list your projects. You should see the tools the connection is allowed to use: the write tools appear only when you granted read and edit, and the doc tools only when documentation access is on.
Removing the connector revokes the access again, and so does revoking it from Agenticboard. If you would rather use a Personal Access Token, any header-capable client takes one — Claude Code, Cursor and Claude Desktop are set up that way. Both paths hit the same MCP server.
What you approved decides the shape of the toolset. These are the tools the MCP server exposes.
list_projects, get_board, get_task and list_members give it columns, cards, labels, comments, attachments and testing feedback. create_task and update_task open a card or change its title, description, status, priority, type, assignee, deadline, PR URL, version or cycle. get_workflow returns the project's ordered steps, and the write tools return the next step to take, so the work happens in the order you set rather than the order it occurs to anyone. find_resolved searches done tasks and returns how each was fixed, and get_task_history gives one ticket's full trail. list_docs, get_doc, create_doc and update_doc work the same pages your team reads. Edits carry a reason into the doc's version history. comment_task, add_test_report and attach_file put the test result and the screenshot on the card, where a reviewer will look for them. search covers projects, tasks and docs, and get_history with search_history reach the knowledge behind them: doc changes, resolved tickets and captured errors. The connection acts as you, narrowed by what you approved, and every call it makes is logged under the token's Activity in Agenticboard.
Set up a board, write the first ticket, and ask Claude what is in flight without leaving the conversation.
MCP access comes with the Pro plan — see pricing.
Other guides: Claude Code · Cursor · Claude Desktop · Sentry