JetBrains IDEs · MCP
Junie does the work.
A person signs it off.
Get the right person's approval before Junie or AI Assistant changes something that is not yours to allow, with the answer kept after the IDE closes. Today the agent asks whoever is at the keyboard; add one MCP server and it asks the person who owns the decision, wherever they are, and carries on when they answer.
No card required · Approvals on every plan · No token to paste
The situation
What happens when JetBrains IDEs needs a person today
The agent asks the person at the keyboard
Junie stops before a command runs and asks whoever is in the IDE. The person who should decide about a migration, a deploy or a message to a customer usually is not.
A click in a panel is not a record
When the change is questioned later, nothing says who allowed it or what they were shown. The decision closed with the tool window.
Brave mode is a decision nobody made
Prompt fatigue ends in letting the agent run without asking. From then on nothing is weighed, and nothing records that the check was switched off.
Setup
Connect JetBrains IDEs
Junie and AI Assistant in IntelliJ IDEA, PyCharm, Rider and the rest ask a named person before they act, and publish what they build to an address.
- 1
For Junie, add it to ~/.junie/mcp/mcp.json
{ "mcpServers": { "deliverd": { "url": "https://deliverd.dev/api/mcp/v2" } } }Or .junie/mcp/mcp.json in a project for that repository only. Keep any servers already listed there.
- 2
Authorise it
Junie lists the server as needing authorisation. Authorise it and approve the connection in your browser.
- 3
For AI Assistant, add it in Settings → Tools → AI Assistant → Model Context Protocol (MCP)
{ "mcpServers": { "deliverd": { "command": "npx", "args": ["-y", "mcp-remote", "https://deliverd.dev/api/mcp/v2"] } } }Click Add and paste this. mcp-remote opens your browser for the approval, so there is still no token to paste. Needs Node.js.
No token is written anywhere. Deliverd implements the MCP authorization spec, so the client registers itself and you approve it in a browser — and removing someone from your organisation cuts their connection off at once, rather than whenever a token happens to expire.
What you get
Once it is connected
Everything below is one call from JetBrains IDEs, over the connection you just made — the ask first, and publishing last.
One call to ask a person
request_approval, request_review or request_information over the connection above. The agent says what it wants to do, why, and what it affects, and the question goes to a named person — not to whoever is at the keyboard.
Decided wherever they are
They approve, correct a figure the agent marked as correctable, reject with a reason or ask the agent a question — on a page built for a phone, from an email link, or from a Slack or Teams card. A colleague can cover while they are away.
The agent carries on
The decision comes back as something the agent can branch on: poll get_approval, wait on a callback URL, or let the SDK block. Approved, it does the action. Rejected, it stops and knows why.
A check before it acts
Before a payment, a deployment or a message to a customer, check_gate declares the action. With no rules a named person decides every one; with rules, the routine cases are settled and the rest go to a person.
A record of who allowed what
Who was asked, what they were shown, what they decided and when — kept as the work happens, and exported as an evidence pack in JSON, CSV or PDF with a digest so an altered copy can be told apart.
And when there is a report, a URL instead of a file
publish_report puts the HTML the agent writes on an address behind your organisation's sign-in, versioned, with comments anchored to the passage they are about. One capability among the rest, over the same connection.

request_approval over the connection above, answered on a page that works on a phone, with the decision back in JetBrains IDEs as soon as it is made. See how approvals work →Questions
JetBrains IDEs and Deliverd.
Do I need a plugin?
No. Junie and AI Assistant already speak MCP, so there is nothing to install from the Marketplace. One entry in their MCP settings connects every JetBrains IDE that runs them: IntelliJ IDEA, PyCharm, WebStorm, Rider, GoLand and the rest.
Where does the config go?
For Junie, ~/.junie/mcp/mcp.json for every project, or .junie/mcp/mcp.json inside one. For AI Assistant, Settings → Tools → AI Assistant → Model Context Protocol (MCP). Merge the entry into what is already there.
Do I need a token?
No. Junie supports OAuth for remote servers: it registers itself and opens a browser for you to approve the connection. For AI Assistant, the mcp-remote bridge does the same sign-in, so the config holds only an address.
What can Junie do once it is connected?
Ask a named person for approval, a review or information it is missing, check a gate before a consequential action, and keep a task's objective and state across sessions — all over the same connection. It can also publish a report, update it in place, read reader comments and resolve them.
Let JetBrains IDEs act. Know who said it could.
One call to ask a person, answered where they are, and a record of who allowed what — from the tool you already have open.