
JFrog: Propagating User Identity From AI Agents to Your Tools
JFrog explains how to propagate end-user identity across the 'second hop' when AI agents call tools through Amazon Bedrock AgentCore Gateway, focusing on On-Behalf-Of (OBO) token exchange to authenticate to JFrog Artifactory as the actual user rather than a shared service credential. It compares OBO against header whitelisting and three-legged OAuth, explaining why shared credentials break audit fidelity, least privilege, and containment for agentic systems, and provides a CLI-driven procedure to configure OBO between AgentCore Identity and Artifactory's MCP server.
TL;DR Walkthrough of using On-Behalf-Of token exchange so AI agents calling JFrog Artifactory via Amazon Bedrock AgentCore Gateway act with per-user identity instead of a shared credential.
Organisations named here
- AgentCore Identity
Technologies named here
- Amazon Bedrock AgentCore Gateway
- JFrog Artifactory
- Amazon CloudWatch
- AWS CLI
- curl
- Model Context Protocol
- OAuth
- AgentCore Identity
- JFrog Access
Techniques named here
- Header whitelisting (header propagation)
- OAuth 2.0 Token Exchange (RFC 8693)
- OIDC identity mapping
- Three-legged OAuth (authorization-code grant)
Concepts named here
- Audit fidelity
- Containment
- Identity propagation
- Least privilege
- On-behalf-of (OBO) token exchange
Related in AI Security
