Synchronize models with AWS Bedrock

Link ​ValidMind models to foundation models and generative AI agents in Amazon Bedrock, supporting governance across your AI/ML ecosystem.

Prerequisites

• You are logged in to ValidMind.
• You are a Customer Admin or assigned another role with sufficient permissions to configure connections.¹
• You have AWS IAM credentials with permissions to access Bedrock.

¹ Manage permissions

AWS Bedrock exposes two primary integration surfaces:

• Foundation model layer — Access to AWS-hosted and custom foundation models through the Bedrock model catalog (Discover and Tune views).

• Application layer — Bedrock primitives such as agents, flows, and knowledge bases that represent built generative AI applications and workflows.

Configure the connection

1. In the left sidebar, click Settings.

2. Under Integrations, select Connections.

3. Click Add Connection.

4. In the modal that opens, select AWS Bedrock.

5. Complete:

   • integration name — How other admins can identify the connection.
   • description (optional) — The intended usage or additional details.
   • aws region — The primary region where your Bedrock resources are deployed, for example us-east-1.
   • aws access key id — The secret generated by AWS IAM with permissions to access Bedrock.
   • aws secret access key — The secret generated by AWS IAM with permissions to access Bedrock.
   • initial status — Set to Operational to enable immediately or Disabled if you plan to finish setup later.

6. Click Save Integration.

7. Test the connection:

   1. Hover over the AWS Bedrock connection you just created.
   2. When the menu appears, click on it and select Test Connection.

   If the test is successful, the message Connection successful displays.

Link models from Bedrock

Once the connection is configured, you can link ​ValidMind models to your Bedrock resources:

NoneModel dependencies

Bedrock agents, flows, and runtimes typically call foundation models (LLMs) under the hood. When you link these application-layer resources to ​ValidMind, consider also registering their underlying foundation models and creating dependency relationships between them.

This builds a complete dependency graph showing how your AI applications relate to underlying models. Benefits include:

• Impact analysis — See which applications are affected if a foundation model is deprecated or has compliance issues.
• Full audit trails — Documentation and risk assessments capture the entire model stack, not just the top-level application.
• Governance visibility — Track model versions and changes across your AI ecosystem.

1. In the left sidebar, click Inventory.

2. Select a model by clicking on it or find your model by applying a filter or searching for it.²

3. Scroll down until you locate the Amazon Bedrock connection box in the right sidebar.

4. Hover over the Amazon Bedrock box.

5. When the menu appears, click on it and select Link Model.

6. In the modal that opens, use the tabs to filter by resource type:

   • All — View all available Bedrock resources.
   • Foundation Models — AWS-hosted and custom foundation models from the Bedrock catalog.
   • Agents — Generative AI agents you’ve built in Bedrock.
   • Flows — Prompt flows and orchestration workflows.
   • AgentCore Runtimes — Agent runtime environments.

7. Click select model to choose the specific Bedrock resource to link.

8. Optional: Click Test Connection to ensure the connection is working as expected.

   If the test is successful, the message Connection Test Successful displays.

9. Click Link Model.

² Working with the inventory