You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
-[A `compose.yaml` file in your project](https://docs.docker.com/compose/gettingstarted/)
16
+
-[A Dockerfile in your project](https://docs.docker.com/get-started/docker-concepts/building-images/writing-a-dockerfile/) or use [Railpack](../concepts/railpack.md)
17
+
18
+
:::warning["MCP prerequisite"]
18
19
If you would like to deploy using [MCP](https://modelcontextprotocol.io/), you like need to setup the [Defang MCP Server](../concepts/mcp.md) with the IDE of your choice.
19
20
:::
20
21
21
22
## CLI
22
23
23
24
### Step 1 - Navigate to your project directory
25
+
24
26
Head to the folder where your project is located.
27
+
25
28
```bash
26
29
$ cd path/to/your/project
27
30
```
28
31
29
-
### Step 2 - Authenticate Defang to use your AWS Account
32
+
### Step 2 - Configure your AWS credentials
30
33
31
-
There are many ways to authenticate your [AWS account](https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-authentication.html).
34
+
There are many ways to authenticate with your [AWS account](https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-authentication.html).
32
35
Once you've done that, Defang will look for your AWS credentials in your shell environment and expect to find one of the following credential sets:
33
-
* the `AWS_PROFILE` environment variable
34
-
* or, the `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, and `AWS_REGION` environment variables
36
+
37
+
- the `AWS_PROFILE` environment variable, and optionally the `AWS_REGION` environment variable
38
+
- or, the `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, and `AWS_REGION` environment variables
35
39
36
40
:::tip
37
41
If you have the AWS CLI installed (which is not required in order to use the Defang CLI), you can verify that you've authenticated to AWS by running `aws sts get-caller-identity` and see your account ID.
@@ -56,29 +60,50 @@ Defang will provision resources in your AWS account and deploy your services. Yo
56
60
## MCP
57
61
58
62
### Step 1 - Navigate to your project directory with your IDE
63
+
59
64
Navigate to your project directory and open it in an IDE that has the [Defang MCP Server](../concepts/mcp.md) installed.
60
65
61
66
:::tip
62
67
If the Defang MCP server doesn’t start automatically, try starting or restarting it from your IDE settings.
63
68
64
69
:::
65
70
66
-
### Step 2 - Setting your provider
67
-
In your chat window, type `/mcp.defang.AWS_Setup` or select the corresponding prompt from your IDE’s options.
71
+
### Step 2 - Configure your AWS credentials
72
+
73
+
There are many ways to authenticate with your [AWS account](https://docs.aws.amazon.com/cli/latest/userguide/cli-chap-authentication.html).
74
+
75
+
:::tip
76
+
If you have the AWS CLI installed (which is not required in order to use the Defang CLI), you can verify that you've authenticated to AWS by running `aws sts get-caller-identity` and see your account ID.
77
+
:::
78
+
79
+
Once you’ve set up your `AWS_PROFILE` or retrieved your `AWS_ACCESS_KEY_ID`, `AWS_SECRET_ACCESS_KEY`, and `AWS_REGION`, you’re ready to continue. Open your chat window and type /defang.AWS_Setup, or select the corresponding prompt from your IDE’s options.
68
80
69
81
70
82
71
83
The [MCP prompt](https://modelcontextprotocol.io/specification/2025-06-18/server/prompts) accepts two types of credentials:
72
-
* the AWS Profile arguement (skip the rest of the prompts after entering the AWS Profile)
73
-
* or, the AWS_ACCESS_KEY_ID, AWS_SECRET_ACCESS_KEY, and AWS_REGION arguments
84
+
85
+
1. Method 1 - Using an AWS Profile:
86
+
The following prompt will have three arguments and only the first argument is required:
87
+
88
+
- AWS Credential - REQUIRED : The name of the AWS profile you want to use (e.g., "default" or "my-profile"). 
89
+
- AWS_SECRET_ACCESS_KEY - IGNORE : Leave this argument blank and press Enter. 
90
+
- AWS_REGION - OPTIONAL : The AWS region where you want to deploy your services (e.g., "us-west-2"). 
91
+
92
+
2. Method 2 - Using Access Keys:
93
+
The following prompt will have three arguments and all three arguments are required:
94
+
95
+
- AWS Credential - REQUIRED : The AWS Access Key ID (e.g., "AKIAIOSFODNN7EXAMPLE"). 
- AWS_REGION - REQUIRED : The AWS region where you want to deploy your services (e.g., "us-west-2"). 
74
98
75
99
### Step 3 - Deploy
76
-
Once the prompt completes without errors, your chat input will be automatically populated with the deployment command for the LLM to execute.
100
+
101
+
Once the prompt completes without errors, your chat input will be automatically populated with the deployment command for the IDE to execute.
77
102
78
103
### Step 4 - Inspect your deployment
79
-
You can check the status of your depolyment by using the [Defang Services tool](../concepts/mcp#services) or you can inspect the resources created in your AWS Dashboard.
80
104
105
+
You can check the status of your deployment by using the [Defang Services tool](../concepts/mcp#services) or you can inspect the resources created in your AWS Dashboard.
81
106
82
-
***
107
+
---
83
108
84
109
For a deeper discussion of the Defang AWS Architecture, including a list of the resources we provision in your account, see our [AWS Provider docs](/docs/providers/aws).
-[A `compose.yaml` file in your project](https://docs.docker.com/compose/gettingstarted/)
17
+
-[A Dockerfile in your project](https://docs.docker.com/get-started/docker-concepts/building-images/writing-a-dockerfile/) or use [Railpack](../concepts/railpack.md)
18
+
19
+
:::warning["MCP prerequisite"]
19
20
If you would like to deploy using [MCP](https://modelcontextprotocol.io/), you like need to setup the [Defang MCP Server](../concepts/mcp.md) with the IDE of your choice.
20
21
:::
21
22
22
23
## CLI
23
24
24
25
### Step 1 - Navigate to your project directory
26
+
25
27
Head to the folder where your project is located.
28
+
26
29
```bash
27
30
$ cd path/to/your/project
28
31
```
29
32
30
-
### Step 2 - Authenticate Defang to use your GCP Account
33
+
### Step 2 - Configure your GCP credentials
31
34
32
35
After signing in to your GCP account, select an existing project or [create a new project](https://developers.google.com/workspace/guides/create-project). Make sure [billing is enabled](https://cloud.google.com/billing/docs/how-to/modify-project). Then, note down the project ID and set it as environment variable `GCP_PROJECT_ID`.
33
36
@@ -50,30 +53,35 @@ Defang will provision resources in your GCP account and deploy your services. Yo
50
53
## MCP
51
54
52
55
### Step 1 - Navigate to your project directory with your IDE
56
+
53
57
Navigate to your project directory and open it in an IDE that has the [Defang MCP Server](../concepts/mcp.md) installed.
54
58
55
59
:::tip
56
60
If the Defang MCP server doesn’t start automatically, try starting or restarting it from your IDE settings.
57
61
58
62
:::
59
63
60
-
### Step 2 - Authenticate Defang to use your GCP Account
64
+
### Step 2 - Configure your GCP credentials
61
65
62
66
After signing in to your GCP account, select an existing project or [create a new project](https://developers.google.com/workspace/guides/create-project). Make sure [billing is enabled](https://cloud.google.com/billing/docs/how-to/modify-project). Then, note down the project ID for the next step.
63
67
64
68
### Step 2 - Setting your provider
65
-
In your chat window, type `/mcp.defang.GCP_Setup` or select the corresponding prompt from your IDE’s options.
69
+
70
+
In your chat window, type `/defang.GCP_Setup` or select the corresponding prompt from your IDE’s options.
66
71
67
72
68
73
69
-
The [MCP prompt](https://modelcontextprotocol.io/specification/2025-06-18/server/prompts) will require your GCP project ID you created in [Step 2](#step-2---authenticate-defang-to-use-your-gcp-account-1).
74
+
The [MCP argument](https://modelcontextprotocol.io/specification/2025-06-18/server/prompts) will require your GCP project ID you created in [Step 2](#step-2---authenticate-defang-to-use-your-gcp-account-1).
Once the prompt completes without errors, your chat input will be automatically populated with the deployment command for the LLM to execute.
78
+
79
+
Once the prompt completes without errors, your chat input will be automatically populated with the deployment command for the IDE to execute.
73
80
74
81
### Step 4 - Inspect your deployment
82
+
75
83
You can check the status of your depolyment by using the [Defang Services tool](../concepts/mcp#services) or you can inspect the resources created in your GCP Dashboard.
76
84
77
-
***
85
+
---
78
86
79
87
For a deeper discussion of the Defang GCP Architecture, see our [GCP docs](/docs/providers/gcp).
-[A `compose.yaml` file in your project](https://docs.docker.com/compose/gettingstarted/)
15
+
-[A Dockerfile in your project](https://docs.docker.com/get-started/docker-concepts/building-images/writing-a-dockerfile/) or use [Railpack](../concepts/railpack.md)
16
16
17
-
:::warning["MCP Pre-requisite"]
17
+
:::warning["MCP prerequisite"]
18
18
If you would like to deploy using [MCP](https://modelcontextprotocol.io/), you like need to setup the [Defang MCP Server](../concepts/mcp.md) with the IDE of your choice.
19
19
:::
20
20
@@ -79,11 +79,11 @@ If the Defang MCP server doesn’t start automatically, try starting or restarti
79
79
80
80
### Step 2 - Setting your provider
81
81
82
-
In your chat window, type `/mcp.defang.Playground_Setup` or select the corresponding prompt from your IDE’s options.
82
+
In your chat window, type `/defang.Playground_Setup` or select the corresponding prompt from your IDE’s options.
0 commit comments