JFrog Introduces AgentSecOps to Secure AI Coding Workflows
Event summary
- JFrog unveiled a new suite of AgentSecOps capabilities at swampUP 2026 to secure and govern AI agent workflows.
- The enhancements include AI asset scanning, an agent plugins registry, and network-layer protection to block risky or malicious models.
- Gartner predicts 40% of agentic AI projects will be canceled by 2027 due to escalating costs or inadequate risk controls.
- JFrog's new features are available immediately to customers.
The big picture
JFrog's move to embed security into AI agent workflows addresses a critical gap in the software supply chain. As AI agents become more prevalent, traditional DevSecOps controls are proving inadequate, with Gartner warning of high project failure rates due to security risks. JFrog's solution aims to provide a trusted system of record for all artifacts consumed or produced by AI agents, ensuring compliance and security from ingestion to delivery. This strategic shift could set a new standard for securing agentic workflows in the enterprise.
What we're watching
- Adoption Pace
- How quickly enterprises will integrate JFrog's AgentSecOps capabilities into their AI agent workflows.
- Security Effectiveness
- Whether JFrog's new security measures will successfully prevent software supply chain attacks in agentic environments.
- Market Differentiation
- The extent to which JFrog can maintain a competitive edge in the AI security space as more players enter the market.
Related topics