How Eon built multi-cloud investigation with an MCP Gateway on Amazon Bedrock AgentCore
Learn how Eon replaced per-tool credentials and local MCP processes across AWS, GCP, and Azure with a single authenticated Amazon Bedrock AgentCore gateway, giving more than 100 developers and autonomous agents unified access to production investigation and cutting new-engineer onboarding from two days to two minutes.
First seen by Oriel on .
AI summary
Eon replaced per-tool credentials and local MCP servers with an Amazon Bedrock AgentCore Gateway, giving developers and agents unified, policy-controlled access to investigation tools across AWS, GCP, and Azure.
- Cedar policies enforce tool-level permissions for human SSO identities and agents, with production tools read-only and selected write actions restricted to explicitly permitted agent identities.
- AgentCore Runtimes host MCP targets across AWS accounts, connecting through IAM role assumption instead of VPC peering or per-account PrivateLink.
- Onboarding fell from two days to about two minutes; developers use zero local MCP processes, and more than 100 developers processed approximately 10,000 sessions.
- DYNAMIC listing reduces Cedar policy size by fetching tool manifests at request time, but disables semantic tool search; Eon combines it with DEFAULT mode for other targets.
- An Amazon API Gateway shim translates OAuth state handling for standard MCP clients, while Runtime wrappers filter large tool catalogs for individual developers.
#amazon bedrock agentcore #mcp #cedar #cloud security
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| AWS Machine Learning Blog | d3c3a7c789540007b485ac6f7137782fb5e6c556 | https://aws.amazon.com/blogs/machine-learning/how-eon-built-multi-cloud-investigation-with-an-mcp-gateway-on-amazon-bedrock-agentcore/ Read discussion |