# CodeRabbit MCP connector

Connect CodeRabbit to Elaichi and Claude, ChatGPT, Cursor, Microsoft Copilot or the Elaichi Agent can read review comments, security findings and review metrics, manage seats, roles and learnings, and run reports, each person inside their own access.

Source: https://elaichi.ai/connectors/coderabbit/

## Facts

| | |
| --- | --- |
| Application | CodeRabbit |
| Category | Product Development |
| AI tools | 21 |
| Authentication | Connects with an API key |
| Bring your own OAuth app | No |
| MCP endpoint | https://api.elaichi.ai/mcp |
| Works with | Claude, ChatGPT, Cursor, Microsoft Copilot, any MCP client, and the Elaichi Agent |
| Tools advertised by name | No. Connected tools are never listed one by one, however few there are. The endpoint advertises `search_tools` and `execute_tool` instead |

## What you can ask once CodeRabbit is connected

- List open security findings across our repositories this week
- Summarize review metrics for the platform team this month
- Which CodeRabbit learnings haven't been updated this quarter?

## Connect CodeRabbit in Elaichi

This happens once for the organization, before any client is involved.

1. Open Connections, choose Add connection, and pick CodeRabbit.
2. Optionally set Share with, then press Connect.
3. Paste a CodeRabbit API key. One person generates a token in CodeRabbit and pastes it once. Everyone else works through Share with, and never sees it.

Credentials are vaulted and nobody, including the AI, reads them back. The connection becomes a toolbox immediately, so you can curate which CodeRabbit tools are exposed, rename them, or freeze arguments before anyone points a client at it.

## CodeRabbit MCP connector for Claude

Endpoint: https://api.elaichi.ai/mcp

1. Open Customize, then Connectors.
2. Press Add.
3. Name it, paste the MCP server URL, then Continue.
4. Sign in and approve.

On Team and Enterprise, an Owner adds it once. Everyone else turns it on for themselves.

## CodeRabbit MCP connector for Claude Code

Endpoint: https://api.elaichi.ai/mcp

1. In a terminal, run `claude mcp add` with the endpoint.
2. Start Claude Code and run `/mcp`.
3. Pick elaichi, then sign in and approve in the browser.

```bash
claude mcp add --transport http elaichi --scope user https://api.elaichi.ai/mcp
```

Use `--scope project` instead to share it with the repository through `.mcp.json`.

## CodeRabbit MCP connector for ChatGPT

Endpoint: https://api.elaichi.ai/mcp

1. Open Plugins, then press the + button.
2. Name it and paste the endpoint into Server URL.
3. Leave Authentication on OAuth, then tick the risk acknowledgement.
4. Press Create, then sign in and approve.

Works on the web. The plugin directory lives at chatgpt.com/plugins.

## CodeRabbit MCP connector for Cursor

Endpoint: https://api.elaichi.ai/mcp

1. Open `~/.cursor/mcp.json`.
2. Add the endpoint under `mcpServers`.
3. Reload Cursor, then sign in and approve.

Set up per machine, so repeat it on each computer you work from.

## CodeRabbit MCP connector for Microsoft Copilot

Endpoint: https://api.elaichi.ai/mcp

1. In Copilot Studio, open your agent, then Tools, Add a tool, New tool and Model Context Protocol.
2. Paste the endpoint into Server URL, choose OAuth 2.0, then Dynamic discovery, and press Create.
3. Create a new connection, sign in and approve, then press Add to agent.
4. Under Channels, open Teams and Microsoft Copilot and make the agent available in Microsoft Copilot.

Microsoft Copilot was called Microsoft 365 Copilot. Keep the tool on end-user credentials, so each person signs in as themselves.

## Connect CodeRabbit to any MCP client

Endpoint: https://api.elaichi.ai/mcp

1. Add the endpoint as a remote MCP server.
2. Sign in and approve.

The Elaichi Agent already has these tools, with nothing to set up.

## What the consent screen decides

Every scope the client asks for starts ticked except "Delete things", which you tick yourself. Calling a CodeRabbit tool needs "Use your connected apps". Over MCP there is no trusted place to confirm a write in the moment, so the consent screen is the standing approval rather than a formality. Think hard before granting Use your connected apps, which can also delete in connected apps and cannot be undone.

## What teams do with CodeRabbit through Elaichi

### See what CodeRabbit flagged this week

Engineering. Ask for the review comments CodeRabbit left across your repositories since Monday and get them grouped by repository, so standup starts with the real list instead of a scroll through pull requests.

### Pull review metrics for the quarterly update

Engineering leadership. Ask for CodeRabbit review metrics by team and period, then have a report created from them, ready to paste into the leadership deck without exporting anything by hand.

### Round up open security findings

Security. Ask which security code findings CodeRabbit has raised across the organization's repositories, sorted by severity, and get a short list of what needs an owner this week.

### Keep CodeRabbit learnings current

Platform. List the learnings CodeRabbit has picked up, update the ones that describe an old convention, and delete the ones that no longer apply, all from one conversation.

### Assign seats and roles in bulk

IT. Hand a list of new engineers to the agent and have their CodeRabbit seats assigned and roles set in one pass, with the seat assignment mode changed when the team's policy changes.

### Answer an audit question from the log

Compliance. Ask who changed a role or seat in CodeRabbit last month and get the answer straight from the CodeRabbit audit log, with dates, instead of chasing an admin for a screenshot.

## Elaichi vs Zapier MCP vs Composio for CodeRabbit

All three can connect CodeRabbit to an AI assistant, and all three have admin controls. They differ in where access lives and how you pay.

| What to check | Elaichi | Zapier MCP | Composio |
| --- | --- | --- | --- |
| Where the AI connects | One address for the whole organization. Endpoint: https://api.elaichi.ai/mcp | A server per member, created at sign-in. | An MCP endpoint per team, or an SDK. |
| Control over CodeRabbit tools | Allow or restrict single CodeRabbit tools, per role or user. | App and action restrictions on the account. | Role permissions, down to the action. |
| Record of calls | One audit entry per CodeRabbit call. | A History tab of tool calls. | A log of every tool call. |
| Single sign-on | SAML or OIDC, plus SCIM, on Gold. | SAML on Enterprise. | SAML and OIDC on Enterprise. |
| Price | Free for one person; $15 per user per month on Gold. | 2 tasks per successful call. | Billed per tool call. |

Sources: Zapier MCP [docs](https://docs.zapier.com/mcp/get-started/quickstart), [security](https://docs.zapier.com/mcp/manage/security), [usage](https://docs.zapier.com/mcp/features/usage); Composio [docs](https://docs.composio.dev/docs/composio-connect), [gateway](https://composio.dev/mcp-gateway), [enterprise](https://composio.dev/enterprise), [pricing](https://composio.dev/pricing). Checked September 2026.

Longer take: [Zapier MCP alternative](/blog/zapier-mcp-alternative/) and [when you don't need an MCP gateway](/blog/when-you-dont-need-an-mcp-gateway/).

## Frequently asked questions

### How do I connect CodeRabbit to Claude?

Two steps. In Elaichi, pick CodeRabbit and paste in a CodeRabbit API key; there is no OAuth application to register and no client ID or secret to generate. Then in Claude open Customize, then Connectors, then Add, and paste https://api.elaichi.ai/mcp. Claude signs you in through Elaichi and CodeRabbit is ready to use.

### Does CodeRabbit work with ChatGPT and Cursor as well as Claude?

Yes. Once CodeRabbit is connected in Elaichi, the same endpoint, https://api.elaichi.ai/mcp, works in Claude, ChatGPT, Cursor, Microsoft Copilot, any other MCP client and the Elaichi Agent. You connect CodeRabbit once and every client gets it.

### What can an AI agent actually do with my CodeRabbit data?

It can read the review comments CodeRabbit left on your pull requests, list security code findings and review metrics across your repositories, and create a report from them. It can also list users, assign seats and change roles in bulk, create and update roles, tidy up learnings, and read the CodeRabbit audit log. Short concrete asks, like "list this week's security findings for the mobile repository", work better than long paragraphs.

### Does connecting CodeRabbit give the AI access to every organization and repository?

Only what the CodeRabbit API key you connected can see, and only for the person who signed in. Each person signs in to Elaichi as themselves, and Elaichi can narrow what they reach in CodeRabbit by role or by person. Elaichi never widens access beyond what the connected CodeRabbit account already has.

### Can my team share one CodeRabbit connection?

Yes. One person connects CodeRabbit with the API key and shares the connection with a team in Elaichi, and nobody else ever handles the key. Each teammate still signs in as themselves, so the audit log names the person who asked for a report or changed a seat, not the person who connected.

### Can I stop an agent from deleting or changing things in CodeRabbit?

Yes. Restrictions in Elaichi work per action, so you can restrict deleting a role, deleting a learning or changing seats in CodeRabbit while leaving reading review comments and findings open, and set that on a role or on one person. A restricted action still shows up in Elaichi's tool search, flagged as restricted, but no client can run it, whatever the prompt says.

### What happens to a CodeRabbit connection when someone leaves?

Offboarding a person in Elaichi ends their access to CodeRabbit through every client at once. A shared CodeRabbit connection keeps working for everyone else on the team. If you want CodeRabbit gone entirely, disconnecting it once in Elaichi removes it from Claude, ChatGPT, Cursor, Microsoft Copilot and every other client at the same time.

### Does the CodeRabbit MCP connector work with Gemini, Codex, Claude Code or other MCP clients?

Yes. CodeRabbit is reached over the same MCP endpoint every client uses, so anything that speaks MCP can call it — Gemini, Codex, Claude Code, Windsurf, Cline, Zed and OpenCode among them — alongside Claude, ChatGPT, Cursor, Microsoft Copilot, and the Elaichi Agent. The tools on offer and the access behind them are identical whichever client asks. Only the setup screen differs.

### Is Elaichi an alternative to Zapier MCP for CodeRabbit?

Yes. Both let Claude, ChatGPT or Cursor use CodeRabbit. Zapier MCP fits a team that already automates in Zapier, since each person signs in and acts as themselves in that account. Elaichi fits when IT wants one address for the whole company, per-tool rules by role, and a record of every CodeRabbit call.

### How is Elaichi different from Composio for CodeRabbit?

Composio gives AI agents tools and sign-in handling across 1,000+ apps, for developers building agents or people using an assistant, billed per tool call. Elaichi gives a company's own people governed access to CodeRabbit: one address, restrictions per role or user, and $15 per user per month. Both have role permissions and a log of every call.

## All 21 CodeRabbit tools

Every tool below is callable through https://api.elaichi.ai/mcp once CodeRabbit is connected, subject to the toolbox it is in and the restrictions on the caller.

- **List all CodeRabbit users** (List). List all users in the CodeRabbit organization, including seat and role assignments for each user. Returns: user_id, cr_user_id, username, email, seat_assigned, role, role_name, role_type, workspace_role, accounts.
- **CodeRabbit seats bulk assign** (Assign). Bulk assign or unassign CodeRabbit seats for up to 500 users. Organization or self-hosted API keys send action (assign|unassign) with user_ids (provider user IDs); workspace API keys send members (by email or cr_user_id). Returns status (success|partial_success|failure) with per-user results.
- **CodeRabbit seats bulk change roles** (Action). Bulk change CodeRabbit roles for up to 500 users. Returns status (success|partial_success|failure) with per-user results. Required: role, user_ids.
- **Create a CodeRabbit report** (Create). Generate a report with CodeRabbit by invoking the report.generate operation. Returns the generated report payload with fields as defined by CodeRabbit's report.generate API; the JSON request body follows the same reference.
- **List all CodeRabbit audit logs** (List). List CodeRabbit audit log entries. Returns: id, created_at, updated_at. Max 100 per page.
- **Delete a CodeRabbit learning by ID** (Delete). Delete a CodeRabbit learning by id; deletion is idempotent and also succeeds when the learning is already absent. Returns an empty 204 response on success. Required: id. Enterprise plan only.
- **List all CodeRabbit learnings** (List). List CodeRabbit learnings for an organization, filterable by repository, author, file path, and usage stats. Returns: id, learning, organization_id, organization_name, repository_id, repository_name, author_id, author_username, file, source_url, pull_request, created_at, updated_at, usage_count, last_used_at. Enterprise plan only.
- **Update a CodeRabbit learning by ID** (Update). Replace the text of a CodeRabbit learning and refresh its search embedding. Enterprise plan and learning:write permission required. Returns the updated learning. Required: id, learning.
- **List all CodeRabbit review comments** (List). List CodeRabbit review comment metrics for merged pull requests, one record per pull request with the CodeRabbit comments posted on it. Returns: pr_url, organization_id, repository_id, author_id, merged_at, coderabbit_comments. Required: start_date, end_date.
- **List all CodeRabbit review metrics** (List). List CodeRabbit PR review metrics for your organization over a date range. Returns per-pull-request records with pr_url, status, merged_at, review_iterations, estimated_review_minutes, and coderabbit_comments (posted vs accepted counts by severity and category). Required: start_date, end_date.
- **List all CodeRabbit role permission options** (List). List the valid resource_id and access_type values for creating or updating Enterprise custom roles in CodeRabbit. Returns: resource_ids, access_types. Required: org_id when authenticating with a workspace-scoped API token.
- **Create a CodeRabbit role** (Create). Create a custom role in CodeRabbit. Returns the created role including its id, name, role_type, and permissions. Required: name.
- **Delete a CodeRabbit role by ID** (Delete). Delete a custom role in CodeRabbit by id. Only Enterprise custom roles that are not assigned to any organization members can be deleted; system roles cannot. Returns an empty 204 response on success. Required: id.
- **Get single CodeRabbit role by ID** (Get). Get details of a built-in or Enterprise custom role in CodeRabbit by id. Returns: id, name, resource_ids, access_types. Required: id.
- **List all CodeRabbit roles** (List). List built-in and Enterprise custom roles for the authenticated CodeRabbit organization. Returns: id, name, role_type, permissions, user_count.
- **Update a CodeRabbit role by ID** (Update). Update a custom role in CodeRabbit by id. Returns the updated role including its id, name, role_type, and permissions. Required: id.
- **List all CodeRabbit mcp metrics** (List). List MCP server usage for merged, reviewed CodeRabbit pull requests in a UTC date window, one record per pull request with the MCP servers and tools used nested under servers. Join to review metrics on pr_url. Counts are best-effort telemetry and access depends on an organization feature flag. Required: start_date, end_date.
- **List all CodeRabbit security code findings** (List). List CodeRabbit Deep Scan code findings for the repositories in scope, from each repository's latest completed scan unless scan_id is given. Defaults to open findings; filter with state, repo_id, org_id. Includes saved reachability evidence (call stacks) and explanations.
- **List all CodeRabbit organizations** (List). List active CodeRabbit organizations. Organization API keys return their bound organization; workspace API keys return every active organization in the workspace. Use the returned id with the organization repositories tool. Enterprise plan only.
- **List all CodeRabbit organization repositories** (List). List the repositories CodeRabbit records as active for an organization. organization_id is the CodeRabbit organization UUID returned by the list organizations tool. Enterprise plan only. Required: organization_id.
- **Create a CodeRabbit seat assignment mode** (Create). Set the organization's CodeRabbit seat assignment mode (automatic or manual). Only fully self-hosted organizations on the Enterprise plan can use this. Returns: mode. Required: mode.
