Merge pull request #23631 from dvdksn/mcp-client-verify
mcp: add instructions for verifying client connections
@@ -4,6 +4,8 @@ linkTitle: Get started
|
||||
description: Learn how to quickly install and use the MCP Toolkit to set up servers and clients.
|
||||
keywords: Docker MCP Toolkit, MCP server, MCP client, AI agents
|
||||
weight: 10
|
||||
params:
|
||||
test_prompt: Use the GitHub MCP server to show me my open pull requests
|
||||
---
|
||||
|
||||
{{< summary-bar feature_name="Docker MCP Toolkit" >}}
|
||||
@@ -23,24 +25,21 @@ Before you begin, make sure you meet the following requirements to get started w
|
||||
3. Select **Enable Docker MCP Toolkit**.
|
||||
4. Select **Apply**.
|
||||
|
||||
There are multiple ways to get started with Docker MCP Toolkit. You can:
|
||||
The **Learning center** in Docker Desktop provides walkthroughs and resources
|
||||
to help you get started with Docker products and features. On the **MCP
|
||||
Toolkit** page, the **Get started** walkthrough that guides you through
|
||||
installing an MCP server, connecting a client, and testing your setup.
|
||||
|
||||
- The **Learning center** in Docker Desktop provides walkthroughs and resources
|
||||
to help you get started with Docker products and features.
|
||||
Alternatively, follow the step-by-step instructions on this page to:
|
||||
|
||||
On the **MCP Toolkit** page, the **Get started** walkthrough that guides you
|
||||
through installing an MCP server, connecting a client, and testing your
|
||||
setup.
|
||||
- [Install MCP servers](#install-mcp-servers)
|
||||
- [Connect clients](#connect-clients)
|
||||
- [Verify connections](#verify-connections)
|
||||
|
||||
- Alternatively, follow the step-by-step instructions on this page to use
|
||||
Docker Desktop or the CLI to [Install an MCP server](#install-an-mcp-server),
|
||||
[add a client, and test your setup](#install-an-mcp-client-and-test-your-setup)
|
||||
with example prompts.
|
||||
## Install MCP servers
|
||||
|
||||
## Install an MCP server
|
||||
|
||||
{{< tabs group="release" >}}
|
||||
{{< tab name="From Docker Desktop">}}
|
||||
{{< tabs >}}
|
||||
{{< tab name="Docker Desktop">}}
|
||||
|
||||
1. In Docker Desktop, select **MCP Toolkit** and select the **Catalog** tab.
|
||||
2. Search for the **GitHub Official** server from the catalog and then select the plus icon to add it.
|
||||
@@ -56,7 +55,7 @@ There are multiple ways to get started with Docker MCP Toolkit. You can:
|
||||
5. Search for the **Playwright** server from the catalog and add it.
|
||||
|
||||
{{< /tab >}}
|
||||
{{< tab name="From the Docker CLI">}}
|
||||
{{< tab name="CLI">}}
|
||||
|
||||
1. Add the GitHub Official MCP server. Run:
|
||||
|
||||
@@ -85,54 +84,325 @@ There are multiple ways to get started with Docker MCP Toolkit. You can:
|
||||
{{< /tab >}}
|
||||
{{< /tabs >}}
|
||||
|
||||
You’ve now successfully added an MCP server. Next, install an MCP client and test your setup with an example prompt.
|
||||
You’ve now successfully added an MCP server. Next, connect an MCP client to use
|
||||
the MCP Toolkit in an AI application.
|
||||
|
||||
## Install an MCP client and test your setup
|
||||
## Connect clients
|
||||
|
||||
After you've installed MCP servers, you can add clients to the MCP Toolkit. These clients can interact with the installed MCP servers, turning the MCP Toolkit into a gateway. In the following section, let’s add the Claude Desktop client.
|
||||
|
||||
{{< tabs group="release" >}}
|
||||
{{< tab name="From Docker Desktop">}}
|
||||
To connect a client to MCP Toolkit:
|
||||
|
||||
1. In Docker Desktop, select **MCP Toolkit** and select the **Clients** tab.
|
||||
2. Find **Claude Desktop** and select **Connect**.
|
||||
2. Find your application in the list.
|
||||
3. Select **Connect** to configure the client.
|
||||
|
||||
If Claude Desktop isn't installed, select **Download** to install it, then select **Connect**.
|
||||
If your client isn't listed, you can connect the MCP Toolkit manually over
|
||||
`stdio` by configuring your client to run the following command:
|
||||
|
||||
3. Restart Claude Desktop if it's running, and it can now access all the servers in the MCP Toolkit.
|
||||
4. Open Claude Desktop and run a test by submitting the following prompt:
|
||||
```plaintext
|
||||
docker mcp gateway run
|
||||
```
|
||||
|
||||
```text
|
||||
Take a screenshot of the header element on docs.docker.com
|
||||
```
|
||||
For example, if your client uses a JSON file to configure MCP servers, you may
|
||||
add an entry like:
|
||||
|
||||
Claude prompts you for permissions and shares a screenshot of the header element from the Docker documentation page.
|
||||
```json {title="Example configuration"
|
||||
{
|
||||
"servers": {
|
||||
"MCP_DOCKER": {
|
||||
"command": "docker",
|
||||
"args": ["mcp", "gateway", "run"],
|
||||
"type": "stdio"
|
||||
}
|
||||
}
|
||||
}
|
||||
```
|
||||
|
||||

|
||||
Consult the documentation of the application you're using for instructions on
|
||||
how to set up MCP servers manually.
|
||||
|
||||
## Verify connections
|
||||
|
||||
Refer to the relevant section for instructions on how to verify that your setup
|
||||
is working:
|
||||
|
||||
- [Claude Code](#claude-code)
|
||||
- [Claude Desktop](#claude-desktop)
|
||||
- [OpenAI Codex](#codex)
|
||||
- [Continue](#continue)
|
||||
- [Cursor](#cursor)
|
||||
- [Gemini](#gemini)
|
||||
- [Goose](#goose)
|
||||
- [Gordon](#gordon)
|
||||
- [LM Studio](#lm-studio)
|
||||
- [OpenCode](#opencode)
|
||||
- [Sema4.ai](#sema4)
|
||||
- [Visual Studio Code](#vscode)
|
||||
- [Zed](#zed)
|
||||
|
||||
### Claude Code
|
||||
|
||||
If you configured the MCP Toolkit for a specific project, navigate to the
|
||||
relevant project directory. Then run `claude mcp list`. The output should show
|
||||
`MCP_DOCKER` with a "connected" status:
|
||||
|
||||
```console
|
||||
$ claude mcp list
|
||||
Checking MCP server health...
|
||||
|
||||
MCP_DOCKER: docker mcp gateway run - ✓ Connected
|
||||
```
|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```console
|
||||
$ claude "{{% param test_prompt %}}"
|
||||
```
|
||||
|
||||
### Claude Desktop
|
||||
|
||||
Restart Claude Desktop and check the **Search and tools** menu in the chat
|
||||
input. You should see the `MCP_DOCKER` server listed and enabled:
|
||||
|
||||

|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```plaintext
|
||||
{{% param test_prompt %}}
|
||||
```
|
||||
|
||||
### Codex
|
||||
|
||||
Run `codex mcp list` to view active MCP servers and their statuses. The
|
||||
`MCP_DOCKER` server should appear in the list with an "enabled" status:
|
||||
|
||||
```console
|
||||
$ codex mcp list
|
||||
Name Command Args Env Cwd Status Auth
|
||||
MCP_DOCKER docker mcp gateway run - - enabled Unsupported
|
||||
```
|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```console
|
||||
$ codex "{{% param test_prompt %}}"
|
||||
```
|
||||
|
||||
### Continue
|
||||
|
||||
Launch the Continue terminal UI by running `cn`. Use the `/mcp` command to view
|
||||
active MCP servers and their statuses. The `MCP_DOCKER` server should appear in
|
||||
the list with a "connected" status:
|
||||
|
||||
```plaintext
|
||||
MCP Servers
|
||||
|
||||
➤ 🟢 MCP_DOCKER (🔧75 📝3)
|
||||
🔄 Restart all servers
|
||||
⏹️ Stop all servers
|
||||
🔍 Explore MCP Servers
|
||||
Back
|
||||
|
||||
↑/↓ to navigate, Enter to select, Esc to go back
|
||||
```
|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```console
|
||||
$ cn "{{% param test_prompt %}}"
|
||||
```
|
||||
|
||||
### Cursor
|
||||
|
||||
Open Cursor. If you configured the MCP Toolkit for a specific project, open the
|
||||
relevant project directory. Then navigate to **Cursor Settings > Tools & MCP**.
|
||||
You should see `MCP_DOCKER` under **Installed MCP Servers**:
|
||||
|
||||

|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```plaintext
|
||||
{{% param test_prompt %}}
|
||||
```
|
||||
|
||||
### Gemini
|
||||
|
||||
Run `gemini mcp list` to view active MCP servers and their statuses. The
|
||||
`MCP_DOCKER` should appear in the list with a "connected" status.
|
||||
|
||||
```console
|
||||
$ gemini mcp list
|
||||
Configured MCP servers:
|
||||
|
||||
✓ MCP_DOCKER: docker mcp gateway run (stdio) - Connected
|
||||
```
|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```console
|
||||
$ gemini "{{% param test_prompt %}}"
|
||||
```
|
||||
|
||||
### Goose
|
||||
|
||||
{{< tabs >}}
|
||||
{{< tab name="Desktop app" >}}
|
||||
|
||||
Open the Goose desktop application and select **Extensions** in the sidebar.
|
||||
Under **Enabled Extensions**, you should see an extension named `Mcpdocker`:
|
||||
|
||||

|
||||
|
||||
{{< /tab >}}
|
||||
{{< tab name="From the Docker CLI">}}
|
||||
{{< tab name="CLI" >}}
|
||||
|
||||
1. In a terminal window, run the following commands to connect the Claude Desktop client:
|
||||
Run `goose info -v` and look for an entry named `mcpdocker` under extensions.
|
||||
The status should show `enabled: true`:
|
||||
|
||||
```console
|
||||
$ docker mcp client connect claude-desktop --global
|
||||
```
|
||||
|
||||
2. Restart Claude Desktop if it's running, and it can now access all the servers in the MCP Toolkit.
|
||||
3. Open Claude Desktop and run a test by submitting the following prompt using the Sonnet 4 model:
|
||||
|
||||
```text
|
||||
Take a screenshot of the header element on docs.docker.com
|
||||
```
|
||||
|
||||
Claude prompts you for permissions and shares a screenshot of the header element from the Docker documentation page.
|
||||
|
||||

|
||||
```console
|
||||
$ goose info -v
|
||||
…
|
||||
mcpdocker:
|
||||
args:
|
||||
- mcp
|
||||
- gateway
|
||||
- run
|
||||
available_tools: []
|
||||
bundled: null
|
||||
cmd: docker
|
||||
description: The Docker MCP Toolkit allows for easy configuration and consumption of MCP servers from the Docker MCP Catalog
|
||||
enabled: true
|
||||
env_keys: []
|
||||
envs: {}
|
||||
name: mcpdocker
|
||||
timeout: 300
|
||||
type: stdio
|
||||
```
|
||||
|
||||
{{< /tab >}}
|
||||
{{< /tabs >}}
|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```plaintext
|
||||
{{% param "test_prompt" %}}
|
||||
```
|
||||
|
||||
### Gordon
|
||||
|
||||
Open the **Ask Gordon** view in Docker Desktop and select the toolbox icon in
|
||||
the chat input area. The **MCP Toolkit** tab shows whether MCP Toolkit is
|
||||
enabled and displays all the provided tools:
|
||||
|
||||

|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers, either directly in Docker Desktop or using the CLI:
|
||||
|
||||
```console
|
||||
$ docker ai "{{% param "test_prompt" %}}"
|
||||
```
|
||||
|
||||
### LM Studio
|
||||
|
||||
Restart LM Studio and start a new chat. Open the integrations menu and look for
|
||||
an entry named `mcp/mcp-docker`. Use the toggle to enable the server:
|
||||
|
||||

|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```plaintext
|
||||
{{% param "test_prompt" %}}
|
||||
```
|
||||
|
||||
### OpenCode
|
||||
|
||||
The OpenCode configuration file (at `~/.config/opencode/opencode.json` by
|
||||
default) contains the setup for MCP Toolkit:
|
||||
|
||||
```json
|
||||
{
|
||||
"mcp": {
|
||||
"MCP_DOCKER": {
|
||||
"type": "local",
|
||||
"command": ["docker", "mcp", "gateway", "run"],
|
||||
"enabled": true
|
||||
}
|
||||
},
|
||||
"$schema": "https://opencode.ai/config.json"
|
||||
}
|
||||
```
|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```console
|
||||
$ opencode "{{% param "test_prompt" %}}"
|
||||
```
|
||||
|
||||
### Sema4.ai Studio {#sema4}
|
||||
|
||||
In Sema4.ai Studio, select **Actions** in the sidebar, then select the **MCP
|
||||
Servers** tab. You should see Docker MCP Toolkit in the list:
|
||||
|
||||

|
||||
|
||||
To use MCP Toolkit with Sema4.ai, add it as an agent action. Find the agent you
|
||||
want to connect to the MCP Toolkit and open the agent editor. Select **Add
|
||||
Action**, enable Docker MCP Toolkit in the list, then save your agent:
|
||||
|
||||

|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```plaintext
|
||||
{{% param test_prompt %}}
|
||||
```
|
||||
|
||||
### Visual Studio Code {#vscode}
|
||||
|
||||
Open Visual Studio Code. If you configured the MCP Toolkit for a specific
|
||||
project, open the relevant project directory. Then open the **Extensions**
|
||||
pane. You should see the `MCP_DOCKER` server listed under installed MCP
|
||||
servers.
|
||||
|
||||

|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```plaintext
|
||||
{{% param test_prompt %}}
|
||||
```
|
||||
|
||||
### Zed
|
||||
|
||||
Launch Zed and open agent settings:
|
||||
|
||||

|
||||
|
||||
Ensure that `MCP_DOCKER` is listed and enabled in the MCP Servers section:
|
||||
|
||||

|
||||
|
||||
Test the connection by submitting a prompt that invokes one of your installed
|
||||
MCP servers:
|
||||
|
||||
```plaintext
|
||||
{{% param test_prompt %}}
|
||||
```
|
||||
|
||||
## Further reading
|
||||
|
||||
- [MCP Toolkit](/manuals/ai/mcp-catalog-and-toolkit/toolkit.md)
|
||||
|
||||
|
After Width: | Height: | Size: 36 KiB |
|
Before Width: | Height: | Size: 135 KiB |
|
After Width: | Height: | Size: 25 KiB |
BIN
content/manuals/ai/mcp-catalog-and-toolkit/images/cursor.avif
Normal file
|
After Width: | Height: | Size: 16 KiB |
BIN
content/manuals/ai/mcp-catalog-and-toolkit/images/goose.avif
Normal file
|
After Width: | Height: | Size: 44 KiB |
BIN
content/manuals/ai/mcp-catalog-and-toolkit/images/lm-studio.avif
Normal file
|
After Width: | Height: | Size: 20 KiB |
|
After Width: | Height: | Size: 24 KiB |
|
After Width: | Height: | Size: 32 KiB |
|
After Width: | Height: | Size: 28 KiB |
|
After Width: | Height: | Size: 31 KiB |
|
After Width: | Height: | Size: 23 KiB |
@@ -371,4 +371,7 @@
|
||||
- /go/dhi-customization/
|
||||
|
||||
"/dhi/how-to/customize/#create-an-oci-artifact-image":
|
||||
- /go/dhi-customization-artifacts/
|
||||
- /go/dhi-customization-artifacts/
|
||||
|
||||
"/ai/mcp-catalog-and-toolkit/get-started/#verify-connections":
|
||||
- /go/mcp-client-verify/
|
||||
|
||||