Skip to content
worth noting AI agents

Eon unified access to incident investigation tools through Amazon Bedrock AgentCore Gateway

only one source so far

Eon replaced credentials for individual tools and local MCP processes with a shared gateway for developers and AI agents. According to the company, the time needed to set up access for new developers fell from two days to approximately two minutes.

In a post on the AWS blog, Eon described deploying a shared gateway that gives developers and AI agents access to tools for investigating operational incidents across AWS, GCP and Azure clouds. The solution, built on Amazon Bedrock AgentCore, replaces separate credentials and local MCP processes. According to Eon, new developers now gain access to all investigation tools in approximately two minutes, down from two days.

Developers using Claude Code, Cursor or OpenAI Codex configure a single MCP URL and sign in through corporate single sign-on. An autonomous agent for operational reliability uses the same gateway with its own identity. Amazon Cognito handles authentication, and the AgentCore Runtime component exposes local MCP servers as remote HTTP endpoints. The gateway connects dozens of MCP servers with hundreds of tools distributed across several AWS accounts.

Cedar policies check every tool call against the identity and its permissions. Read access is available to selected groups; production tools allow read-only access. Write access is restricted to lower-risk actions, such as commenting on pull requests or work tickets, and requires explicit permission for specific agent identities. The team manages policies as code through controlled changes in GitHub, giving permission changes a traceable history.

What changed

Why it matters

Investigating a single incident may require simultaneous access to container logs, operational metrics, code changes and the status of cloud services. The architecture described shows how to make these tools available to people and agents through a shared entry point while enforcing different permissions for every call. The reported reduction in access setup time is a result achieved by Eon, rather than an independently verified benefit for other teams.

Two audiences, two different impacts

What this means

01

For individuals

Developers at Eon no longer need to configure local MCP servers and separate credentials to gain access to investigation tools; a single MCP URL and a corporate login are enough.

More practical updates →
02

For a business

Eon centralized access management for people and agents in a shared gateway. Role-based permissions, separation of read and write access, and controlled policy changes provide a unified access management process and an audit trail.

Processes
What to decide When designing a similar solution, verify that the gateway policies restrict the tools available to each identity and keep production access read-only.
More business impacts →
Amazon Bedrock AgentCore CEDAR Eon MCP

Check the original

Event sources

only one source so far · 1 publisher, 0 independent. We count feeds from the same owner only once.

1
AWS Machine Learning Blog primary source · first detected How Eon built multi-cloud investigation with an MCP Gateway on Amazon Bedrock AgentCore