Securing AI agent identity with X.509 PKI on AWS
Web page · https://aws.amazon.com/blogs/machine-learning/securing-ai-agent-identity-with-x-509-pki-on-aws
Learn how to secure multi-agent AI systems with cryptographic identity using Keyfactor EJBCA and AWS services. This reference architecture combines signed request envelopes, user-signed delegation assertions, RFC 8693 audience-scoped tokens, and real-time OCSP revocation to prove agent identity and prevent request forgery across agent-to-agent and MCP tool calls.
First seen by Oriel on .
AI summary
The post presents an AWS reference architecture that combines Keyfactor EJBCA and X.509 certificates to authenticate AI agents and authorize their access to MCP tools.
- It uses Amazon Bedrock with the Strands Agents SDK, Keycloak, EJBCA, AWS Secrets Manager, and Amazon RDS for MySQL; the pattern is intended for other MCP-compatible tool servers.
- Three trust layers cover user authentication, agent-to-agent calls, and agent-to-tool access, using signed envelopes, user-signed delegation, audience-scoped RFC 8693 tokens, and OCSP checks.
- Envelopes include a signature, timestamp, and random nonce; receivers check proof of possession, freshness, replay, CA trust, caller role, and certificate revocation.
- The MCP server checks token audience and activity, envelope validity, delegation authorization, and consistency between token, user, and agent identities before executing tools.
- Private keys are generated locally and retained by the application; EJBCA signs CSRs, while issued certificates and keys are stored in AWS Secrets Manager.
#x.509 pki #ai agents #model context protocol #aws
Generated by gpt-5.6-sol. Check the original for details.
Aliases
| Source | External ID | URL |
|---|---|---|
| AWS Machine Learning Blog | 05547fd2bd22554636c30cf4bb67047887b0d94a | https://aws.amazon.com/blogs/machine-learning/securing-ai-agent-identity-with-x-509-pki-on-aws/ Read discussion |