RabbitMQ MCP Server
Provides tools to manage RabbitMQ message brokers, including operations on queues, exchanges, bindings, and other administrative tasks via RabbitMQ's admin API.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@RabbitMQ MCP ServerList all queues on the default vhost"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcp-server-rabbitmq
A Model Context Protocol (MCP) server for RabbitMQ broker management and operations. It lets AI agents manage RabbitMQ brokers conversationally: multi-broker connections, blue-green migration, health checks, and full observability.
Package: amq-mcp-server-rabbitmq on PyPI · Stack: Python, FastMCP, uv
Features
31 tools in v4 (enum-based dispatchers), or 61 tools in v3 (one tool per operation) for broker management - connections, queues, exchanges, health checks, observability, and blue-green migration
16 composable skills - topology visualization, dead letter analysis, capacity planning, and more
Multi-broker support - connect multiple brokers simultaneously, switch between them by alias
Mutative tools gated behind
--allow-mutative-toolsflag (off by default for safety)Tool groups - load only the tools you need with
--tool-groupsSecurity hardened - SSRF protection, credential stripping, TLS warnings, JWKS HTTPS enforcement
Related MCP server: kafka-mcp
Versions
The server ships two tool layouts so you can choose your upgrade path:
Version | Layout | Upgrade impact | Choose it when |
v3 ( | 61 tools, one per operation (e.g. | Non-breaking - a drop-in upgrade from v2.x. | You have existing prompts/integrations bound to v2/v3 tool names and want zero churn. |
v4 ( | 31 enum-based dispatchers (e.g. | Breaking - tool names change (opt in with | You want ~60% fewer tool-description tokens per turn, faster startup, and per-group tool loading (see Why opt in to v4). |
Both are published from this repo. v4 defaults to v3 behavior unless you pass --v4, so installing the latest package never breaks an existing setup on its own. See v4 Mode for the full mapping and the CHANGELOG for the v3 → v4 tool-name table.
Quick Start
Install via PyPI
pip install amq-mcp-server-rabbitmq
# or
uv pip install amq-mcp-server-rabbitmqConfigure in Claude Desktop (v4 mode - recommended)
{
"mcpServers": {
"rabbitmq": {
"command": "uvx",
"args": [
"amq-mcp-server-rabbitmq@latest",
"--v4",
"--allow-mutative-tools"
]
}
}
}Configure in Claude Desktop (v3 mode - legacy, 61 tools)
{
"mcpServers": {
"rabbitmq": {
"command": "uvx",
"args": [
"amq-mcp-server-rabbitmq@latest",
"--allow-mutative-tools"
]
}
}
}Basic Usage
Once configured, the agent can connect to your broker and manage it conversationally:
You: Connect to my RabbitMQ broker at rabbitmq.example.com with user admin
You: List all queues and show me which ones have messages backing up
You: Create a dead letter exchange and bind it to the orders queuev4 Mode
v4 consolidates the 61 individual v3 tools into 31 enum-based dispatchers (29 consolidated groups plus the 2 standalone mutative tools below), reducing context window pressure while preserving full functionality. Each consolidated tool accepts an action parameter to select the operation.
Why opt in to v4
Fewer tokens. Every tool's name and description is sent to the model on every request, whether or not it's used. v4 roughly halves the tool count (61 → 31) and cuts that idle tool-description text by about 60% - from ~2,300 tokens to ~900 tokens (name + docstrings). That is context budget returned to the actual conversation on every single turn, and a smaller tool list also makes the model faster and more accurate at picking the right tool.
Faster startup. Fewer tools means less to register and advertise during the MCP handshake, so the server connects and the client finishes tool discovery sooner. With
--tool-groupsyou can trim further - e.g. load onlyread/healthfor a monitoring agent - registering just the tools that session needs.
Numbers above are measured against this repo's tool definitions; exact token counts vary by client and model tokenizer.
Key Differences from v3
Aspect | v3 | v4 |
Tool count | 61 | 31 (29 dispatchers + 2 standalone mutative) |
Naming |
|
|
Loading | All or nothing | Selectable via |
Compat | N/A |
|
Standalone Mutative Tools
Two tools remain standalone because they are high-impact operations that benefit from explicit invocation and additional confirmation:
close_connection- Close a specific connection by namerebalance_queues- Rebalance queue leaders across cluster nodes
These require the mutative tool group to be loaded.
Configuration
CLI Arguments
Argument | Description |
| Enable v4 consolidated tool mode (31 tools instead of 61) |
| Select which tool groups to load (space-separated). Options: core, read, mutative, migration, observability, health |
| Register v3 tool name aliases alongside v4 tools (for migration) |
| Enable tools that can create, modify, or delete resources (default: off) |
| RabbitMQ Management API port (default: 443 for TLS, 15672 for non-TLS) |
| Use Streamable HTTP transport instead of stdio |
| Port to run the MCP server on (default: 8888) |
| JWKS URI for Bearer Auth Provider |
| Issuer for Bearer Auth Provider |
| Audience for Bearer Auth Provider |
| Required scopes for Bearer Auth Provider |
Tool Groups (v4)
Group | Tools | Gate |
core | connect, connect_oauth, broker, get_skill, get_guideline | Always loaded |
read | queues, exchanges, connections, cluster, entities, overview, policies, shovels, auth | Always loaded |
mutative | manage_queue, manage_exchange, manage_binding, manage_policy, manage_vhost, publish, close_connection, rebalance_queues | Requires |
migration | definitions_export, definitions_import, definitions_compare, definitions_migrate, migration_readiness, federation_setup | Requires |
observability | find_queues | Always loaded |
health | health | Always loaded |
Environment Variables
Variable | Description |
| Log level: DEBUG, INFO, WARNING (default), ERROR |
Tools (v3 layout)
The following table shows the v3 tool names. In v4 mode, these are consolidated into 31 enum-based dispatchers (see the v4 Mode section above). Use --v1-compat to register these names alongside v4 tools.
Connection and Session (6 tools)
Tool | Description |
| Connect to a broker with username/password (SIMPLE auth) |
| Connect to a broker using an OAuth token |
| Switch the active broker by alias |
| List all registered broker connections and which is active |
| Get best-practice guidelines for RabbitMQ deployment and sizing |
| Get a composable workflow recipe (see Skills section below) |
Read-Only: Queues and Exchanges (7 tools)
Tool | Description |
| List all queues in the broker |
| List all exchanges in the broker |
| List all virtual hosts |
| Get detailed information about a specific queue |
| Get detailed information about a specific exchange |
| Get bindings, optionally filtered by queue or exchange |
| Peek at messages in a queue without consuming them |
Read-Only: Connections and Consumers (4 tools)
Tool | Description |
| List all connections on the broker |
| List all consumers on the broker |
| List all open channels on the broker |
| List all users on the broker |
Read-Only: Cluster and Nodes (2 tools)
Tool | Description |
| Get the list of nodes and their info in the cluster |
| Get detailed node info including memory, disk, uptime |
Read-Only: Policies and Shovels (4 tools)
Tool | Description |
| List all policies in a virtual host |
| Get a specific policy by name |
| List all shovels on the broker |
| Get detailed information about a specific shovel |
Read-Only: Permissions and Identity (2 tools)
Tool | Description |
| Get permissions for a user in a virtual host |
| Get the current authenticated user |
Read-Only: Definitions and Migration (3 tools)
Tool | Description |
| Get full broker definitions (exchanges, queues, bindings, users, etc.) |
| Compare definitions between two connected brokers |
| Pre-flight check for blue-green migration |
Read-Only: Observability (3 tools)
Tool | Description |
| Get cluster-wide stats: version, node count, message rates, object totals |
| Find queues by criteria: depth, idle time, no consumers, unacked messages |
| Get connection/channel open and close rates |
Health Checks (8 tools)
Tool | Description |
| Check if the broker is in alarm |
| Check if quorum queues have minimum online quorum |
| Check for local alarms on the active broker |
| Check if TLS certificates expire within a timeframe |
| Check if a protocol listener is active |
| Check health of all virtual hosts |
| List all feature flags and their status |
| List deprecated features currently in use |
Mutative: CRUD (16 tools)
Tool | Description |
| Create a queue (quorum, classic, or stream) |
| Delete a specific queue |
| Remove all messages from a queue |
| Create an exchange (direct, fanout, topic, headers) |
| Delete a specific exchange |
| Create a binding from an exchange to a queue |
| Delete a binding |
| Create or update a policy |
| Delete a policy |
| Create a virtual host |
| Delete a virtual host and all its resources |
| Set permissions for a user in a virtual host |
| Close a specific connection by name |
| Update server definitions |
| Export definitions with optional transformations |
| Import definitions to the active broker |
Mutative: Migration (4 tools)
Tool | Description |
| Export, transform, and import definitions between brokers |
| Set up federation upstream and policy for message draining |
| Rebalance queue leaders across cluster nodes |
| Publish a message via HTTP Management API (diagnostics) |
Mutative: AMQP (2 tools)
Tool | Description |
| Publish a message to a specific queue via AMQP |
| Publish a message to a fanout exchange via AMQP |
Skills
Skills are composable workflows accessed via rabbitmq_broker_get_skill. They guide the agent through multi-step operations by orchestrating existing tools - no additional code required.
Skill | What it does | Tools it composes |
| Check alarms on both brokers + compare definitions for go/no-go | is_in_alarm, compare_definitions |
| Export from source with transforms, import to target | export_definitions, import_definitions |
| Verify federation plugin, create upstream and policy | get_broker_overview, import_definitions |
| Interpret publish/deliver rates and backlog trends | get_queue_info |
| Calculate memory %, disk headroom, FD usage per node | get_node_information, get_cluster_nodes_info |
| Generate Mermaid diagram of exchange-binding-queue graph | list_exchanges, list_queues, get_bindings |
| Predict which queues receive a message given exchange + routing key | get_exchange_info, get_bindings |
| Find queues with no bindings and no consumers | list_queues, get_bindings, list_consumers |
| Find exchanges with no outbound bindings (excluding amq.*) | list_exchanges, get_bindings |
| Walk x-dead-letter-exchange arguments to map the full DLX chain | get_queue_info, get_bindings |
| Peek at DLQ messages and extract x-death headers (source, reason) | get_messages |
| Aggregate dead letters by source queue and rejection reason | list_queues, get_queue_info, get_messages |
| Compare live broker state against best-practice guidelines | get_broker_overview, get_cluster_nodes_info, list_queues, get_guideline |
| Assess queue type, consumers, depth, and policy coverage | get_queue_info, get_guideline |
| Compute resource utilization % vs watermarks, project time-to-alarm | get_cluster_nodes_info |
| Find overlapping policy patterns and report priority winners | list_policies |
Documentation
In-depth guides live in docs/:
Multi-Broker — register and switch between brokers by alias
Authentication — broker auth (SIMPLE/OAuth) and HTTP JWT/JWKS Bearer auth
Skills — the 16 composable workflow recipes
Migration — blue-green migration, definition transforms, federation draining
CHANGELOG — version history and the full v3 → v4 tool-name mapping
Development
Prerequisites
Python 3.10+
uv package manager
Running Locally
git clone https://github.com/amazon-mq/mcp-server-rabbitmq.git
cd mcp-server-rabbitmq
uv run amq-mcp-server-rabbitmqTesting
uv run pytestLinting
uv run ruff check . && uv run ruff format .Docker (Local RabbitMQ)
docker-compose upThis starts RabbitMQ 4 with the management plugin on localhost:5672 (AMQP) and localhost:15672 (Management UI). Default credentials: guest / guest.
Security
Mutative tools disabled by default - pass
--allow-mutative-toolsto opt inMutative action gating - high-impact operations (close_connection, rebalance_queues, set_permissions) require the mutative tool group to be loaded
TLS by default - connections use
use_tls=Trueand port 5671 unless overriddenTLS warnings - non-TLS connections emit a warning in the response so agents can inform users
SSRF protection - hostname validation blocks connections to private/reserved IP ranges and localhost
Credential stripping - definition exports automatically strip passwords and sensitive keys before returning to the agent
URL encoding - all user-supplied names (queues, exchanges, vhosts) are URL-encoded before API calls to prevent injection
JWKS HTTPS enforcement - the
--http-auth-jwks-uriflag rejects non-HTTPS URIs to prevent token validation bypassOAuth support - connect with access tokens instead of username/password
HTTP transport with JWKS auth - run as a remote server with Bearer token validation via configurable IdP
License
Apache-2.0 - see LICENSE for details.
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Tools
- rabbitmq_broker_check_certificate_expirationB
- rabbitmq_broker_check_local_alarmsA
- rabbitmq_broker_check_migration_readinessB
- rabbitmq_broker_check_protocol_listenerC
- rabbitmq_broker_check_virtual_hostsA
- rabbitmq_broker_compare_definitionsA
- rabbitmq_broker_find_queues_by_thresholdB
- rabbitmq_broker_get_bindingsA
Related MCP Servers
- AlicenseBqualityDmaintenanceAn MCP server for RabbitMQ that exposes the RabbitMQ HTTP API as tools, enabling natural language interaction with RabbitMQ management.Last updated1001921MIT
- FlicenseCqualityBmaintenanceExposes Kafka administration operations as MCP tools, enabling AI agents to inspect Kafka clusters using natural language.Last updated1
- Alicense-qualityBmaintenanceEnables LLM agents to manage RabbitMQ queues, exchanges, and shovels via MCP tools, with restricted mode for production environments.Last updatedApache 2.0
- Flicense-qualityDmaintenanceEnables management of RabbitMQ clusters, queues, exchanges, bindings, users, and messages through natural language in chat interfaces, using the MCP protocol.Last updated1
Related MCP Connectors
Build, validate, and deploy multi-agent AI solutions from any AI environment.
Phone, SMS & email for AI agents — one remote MCP endpoint, OAuth login, zero install.
Persistent memory and knowledge management for AI agents with semantic search and 50+ tools.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/amazon-mq/mcp-server-rabbitmq'
If you have feedback or need assistance with the MCP directory API, please join our Discord server