# Use the SDK

Use the CloudSigma AI Studio SDK to manage workspace resources from Go or Python applications. After completing this page, you will use a personal access token to list visible workspaces and obtain the workspace ID required for later tasks.

For all scenarios and task guides, see [AI Studio SDK](https://omnifabric.cloudsigma.com/docs/developer/sdk/ai-studio/index.html.md).

## Prerequisites: Get the SDK

Before starting, go to [Download AI Studio SDK](https://omnifabric.cloudsigma.com/docs/developer/quickstart/downloads/ai-studio.html.md) to obtain the offline distribution package for your language (Python wheel or Go source module), then install and verify dependencies locally.

## 1. Configure the regional endpoint and credentials

Select the API endpoint for the region containing your workspace. The endpoint passed to the SDK must include `/v5`.

| Region | API base URL |
| --- | --- |
| Manila 2, Philippines (`mnl2`) | `https://genai.mnl2.cloudsigma.com/v5` |
| Johannesburg, South Africa (`jhb`) | `https://genai.jhb.cloudsigma.com/v5` |

Create a personal access token in [Access credentials](../../../guides/billing/credentials.md#personal-access-token) in the same regional console. The endpoint, token, workspace ID, and resource IDs must belong to the same region. See [Endpoints & Authentication](https://omnifabric.cloudsigma.com/docs/developer/endpoints-and-authentication.html.md) for details.

The following example uses Manila 2. The packaged examples and SDK task guides read `PRODUCT_API_BASE_URL` and `PRODUCT_API_KEY`:

```bash
export PRODUCT_API_BASE_URL='https://genai.mnl2.cloudsigma.com/v5'
export PRODUCT_API_KEY='<personal-access-token-from-this-region>'
```

If you already configured `CLOUDSIGMA_API_BASE` and `CLOUDSIGMA_API_KEY` using the endpoint guide, reuse them:

```bash
export PRODUCT_API_BASE_URL="$CLOUDSIGMA_API_BASE"
export PRODUCT_API_KEY="$CLOUDSIGMA_API_KEY"
```

The SDK sends the token in `X-API-Key`. Pass its original value without a `Bearer` prefix. When switching regions, switch the endpoint, token, and workspace ID together.

## 2. List visible workspaces

AI Studio SDK features must be used in a workspace. Before using workflows, knowledge bases, data, agents, or other features, select the target workspace and get its workspace ID.

:::::{tab-set}
:sync-group: sdk-language

::::{tab-item} Python
:sync: python

```python
import os

import moi_product_sdk as sdk

client = sdk.new_with_personal_access_token(
    os.environ["PRODUCT_API_BASE_URL"],
    os.environ["PRODUCT_API_KEY"],
)
result = client.workspaces().list()
for workspace in result.workspaces:
    print(workspace.id, workspace.name)
```

::::

::::{tab-item} Go
:sync: go

```go
package main

import (
	"context"
	"fmt"
	"os"

	sdk "github.com/matrixorigin/matrixflow/sdk/go-sdk"
)

func main() {
	ctx := context.Background()
	client, err := sdk.NewWithPersonalAccessToken(
		os.Getenv("PRODUCT_API_BASE_URL"),
		os.Getenv("PRODUCT_API_KEY"),
	)
	if err != nil {
		panic(err)
	}
	result, err := client.Workspaces().List(ctx)
	if err != nil {
		panic(err)
	}
	for _, workspace := range result.GetWorkspaces() {
		fmt.Printf("%s\t%s\n", workspace.GetId(), workspace.GetName())
	}
}
```

::::

:::::

Select a target workspace ID from the output and save it as an environment variable:

```bash
export WORKSPACE_ID='<workspace-id>'
```

## 3. Call example

This example calls the List workflows feature. Use the connection details and workspace ID from the previous steps to create a workspace client and list the workflows that the current identity can read:

:::::{tab-set}
:sync-group: sdk-language

::::{tab-item} Python
:sync: python

```python
import os

import moi_product_sdk as sdk

client = sdk.new_with_personal_access_token(
    os.environ["PRODUCT_API_BASE_URL"],
    os.environ["PRODUCT_API_KEY"],
)
workspace = client.workspace(os.environ["WORKSPACE_ID"])
result = workspace.workflows().list()

print(result.total)
for workflow in result.workflows:
    print(workflow.id, workflow.name, workflow.status)
```

::::

::::{tab-item} Go
:sync: go

```go
package main

import (
	"context"
	"fmt"
	"os"

	sdk "github.com/matrixorigin/matrixflow/sdk/go-sdk"
)

func main() {
	ctx := context.Background()
	client, err := sdk.NewWithPersonalAccessToken(
		os.Getenv("PRODUCT_API_BASE_URL"),
		os.Getenv("PRODUCT_API_KEY"),
	)
	if err != nil {
		panic(err)
	}
	workspace, err := client.Workspace(os.Getenv("WORKSPACE_ID"))
	if err != nil {
		panic(err)
	}
	result, err := workspace.Workflows().List(ctx)
	if err != nil {
		panic(err)
	}

	fmt.Println(result.GetTotal())
	for _, workflow := range result.GetWorkflows() {
		fmt.Printf("%s\t%s\t%s\n", workflow.GetId(), workflow.GetName(), workflow.GetStatus())
	}
}
```

::::

:::::

On success, the program prints the workflow total followed by each workflow's ID, name, and status.

If you receive `401`, check that the token is complete and has no `Bearer` prefix. If you receive `404` or HTML content, check that the endpoint matches the workspace region and includes `/v5`. If the list is empty or you receive `403`, confirm that the current account has the required workspace member or role permissions.

## Next steps

- [Use the AI Studio SDK by scenario](https://omnifabric.cloudsigma.com/docs/developer/sdk/ai-studio/index.html.md)