Skip to content
Onplana Docs

Connect Onplana from the Claude Connectors Directory

All plans Member

Onplana is listed in Anthropic’s official Connectors Directory, so claude.ai and Claude Desktop connect with a browser sign-in and one approval. No personal access token, no JSON config to maintain. Once approved, Claude joins your workspace as an agent persona that can read your projects, work the tasks you assign it, file issues, and build plans.

For the command line, Claude Code uses the manual token and MCP config path in Connect an external agent.

What you get

Section titled “What you get”

Once connected, Claude works with the same Onplana data you can see: list and update tasks, create projects and full plans, read documents, file issues and ideas, post comments, and run reports. It appears in your workspace as a clearly badged agent persona, so everything it does is attributed to it, never silently to you.

Add Onplana to Claude

Section titled “Add Onplana to Claude”

  1. In claude.ai (or Claude Desktop), open your settings and find the Connectors section (the label moves between Claude releases), then browse the directory.

  2. Find Onplana and choose the option to connect it.

  3. A browser window opens on Onplana’s sign-in. Sign in (or pick your account if you already have a session) and review the consent screen, which lists exactly what the connection may do.

  4. Approve. Claude shows the connector as active, and Onplana’s tools are available in your conversations from then on.

Verify the connection

Section titled “Verify the connection”

Ask Claude something only your workspace can answer, for example “what is overdue across my Onplana projects this week?”. A working connection answers with your real tasks. On the Onplana side, Settings → Agents shows the new connection with its last-used time.

Permissions and what Claude can see

Section titled “Permissions and what Claude can see”

The connection inherits your permissions in Onplana, no more, no less:

  • Claude sees only projects you are a member of (or all projects if you are an organization admin), and acts on plan-gated features only at your organization’s tier.
  • It cannot escalate privileges, mint API tokens, change your password, or read security settings. Those surfaces are off the agent path by design.
  • Your conversations live with Anthropic under your Claude account’s data settings, so Onplana data quoted in a chat (project names, task titles, comments) becomes part of that chat history. If that matters for your data policy, treat the connector like any other third-party SaaS handling that data.

Managing the connection

Section titled “Managing the connection”
  • From Claude: remove or disable the Onplana connector in the same connectors settings you added it from. Onplana revokes the token immediately.
  • From Onplana: open Settings → Agents to see every active connection, when it last called, and a revoke button. An organization admin can revoke any member’s connection from the same panel.
  • Token lifetime: Onplana’s token lasts 90 days. When it expires, Claude asks you to sign in again; there is nothing to rotate by hand.

FAQs

Section titled “FAQs”

Do I need a paid Onplana plan? No. The connector works on every plan, including Free. What scales with tier is how many agent connections can be active at once (Free and Starter 2, Pro 3, Business 5, Enterprise 10, Enterprise+ unlimited) and the AI token balance that powers AI-assisted features.

Is this the same as connecting Claude Code? Same server, different doorway. Claude Code connects through the manual MCP path; the directory listing is the route for claude.ai and Claude Desktop. You can use both at once, and they coordinate through execution leases.

Do I ever paste a token? Not on this path. The browser sign-in mints and stores the credential itself. Personal access tokens remain available for the manual path and for CI-style service accounts.

Does the connection use up a seat? No. Claude appears as an agent persona, which is excluded from seat billing. Each teammate connects individually, so every connection acts with that person’s own permissions and is individually revocable.

Section titled “Related”