MODEL HOSTING / Amazon Web Services
AWS Bedrock Custom Model Import MCP Server
Creates and tracks Amazon Bedrock custom model import jobs, and lists, inspects or deletes imported models, from an MCP client. Read-only unless started with --allow-write.
“The Bedrock Custom Model Import Model Context Protocol (MCP) Server streamlines the process of importing custom models into Amazon Bedrock.”
01 / THE REASONING
Why this made the selection.
- Published by AWS in awslabs/mcp; the README at this commit documents six tools: create_model_import_job, list_model_import_jobs, get_model_import_job, list_imported_models, get_imported_model and delete_imported_model.
- The README says the server runs in read-only mode by default and write access must be enabled explicitly, and it names the S3 bucket and IAM role variables (BEDROCK_MODEL_IMPORT_S3_BUCKET, BEDROCK_MODEL_IMPORT_ROLE_ARN) an import uses.
02 / THE REVIEW RECORD
What we actually inspected.
Source review has boundaries.
A clear record is more useful than a “safe” badge.
Material inspected
- src/aws-bedrock-custom-model-import-mcp-server/README.md at e1abc4b3b16a6da4c86ccbfc89de7bd864c5679e
- src/aws-bedrock-custom-model-import-mcp-server/pyproject.toml at the same commit
- src/aws-bedrock-custom-model-import-mcp-server/LICENSE at the same commit
Our findings
- The README's security section says created resources are tagged for traceability and recommends separate IAM roles for read-only and write operations and CloudTrail auditing.
- If BEDROCK_MODEL_IMPORT_S3_BUCKET is set, the server searches that bucket for model files by model name; otherwise create_model_import_job needs a modelDataSource S3 URI.
- pyproject.toml names awslabs.aws-bedrock-custom-model-import-mcp-server 0.0.23, license Apache-2.0.
Not established by this review
- uvx launch
- Any Bedrock model import job
- delete_imported_model
The review applies to the material and revision named here. A newer upstream release can change its behavior.
03 / PUT IT TO WORK
Connect a server deliberately.
- The README gives a client configuration block, not a one-line command: command uvx with args awslabs.aws-bedrock-custom-model-import-mcp-server@latest and, for writes, --allow-write. It was not run.
- Set AWS_PROFILE, AWS_REGION and BEDROCK_MODEL_IMPORT_S3_BUCKET, and BEDROCK_MODEL_IMPORT_ROLE_ARN if the import should use a dedicated role.
Before you start
- uv and Python
- AWS credentials with Amazon Bedrock model-import permissions
- Model files in an S3 bucket
Documented configuration example
Use the client and authentication context described in the setup guidance. Configuration formats vary between hosts.
{
"mcpServers": {
"awslabs.aws-bedrock-custom-model-import-mcp-server": {
"command": "uvx",
"args": [
"awslabs.aws-bedrock-custom-model-import-mcp-server@latest",
"--allow-write"
],
"env": {
"AWS_PROFILE": "your-aws-profile",
"AWS_REGION": "us-east-1",
"BEDROCK_MODEL_IMPORT_S3_BUCKET": "your-model-bucket",
"BEDROCK_MODEL_IMPORT_ROLE_ARN": "your-role-arn"
},
"disabled": false,
"autoApprove": []
}
}
} 04 / FOLLOW THE EVIDENCE
The source trail.
Our notes are separate from the original resource.
Check upstream before adopting a new version.
https://github.com/awslabs/mcp/blob/e1abc4b3b16a6da4c86ccbfc89de7bd864c5679e/src/aws-bedrock-custom-model-import-mcp-server/README.md
Supports: Tools, Read-only default and --allow-write, Environment variables, Security considerations
https://github.com/awslabs/mcp/blob/e1abc4b3b16a6da4c86ccbfc89de7bd864c5679e/src/aws-bedrock-custom-model-import-mcp-server/LICENSE
Supports: Apache License 2.0
https://github.com/awslabs/mcp/blob/e1abc4b3b16a6da4c86ccbfc89de7bd864c5679e/src/aws-bedrock-custom-model-import-mcp-server/pyproject.toml
Supports: Package version 0.0.23, Python >=3.10