The newly introduced capabilities are designed to further transform the way IT teams manage and secure large, complex, and distributed infrastructures.
Cisco announced new innovations to its AgenticOps framework to extend agent-driven operations across network, security, and observability. First introduced last year, AgenticOps is an agent-first IT operating model designed to enable autonomous actions with built-in monitoring. The newly introduced capabilities are designed to further transform the way IT teams manage and secure large, complex, and distributed infrastructures.
“For teams responsible for the operations and security of distributed networks and infrastructure, AgentOps represents a fundamental and fundamental shift away from complexity. This is the true power of Cisco as a platform. By delivering agent capabilities aligned to the key priorities of IT operations, we are empowering teams with Cisco’s unique combination of cross-domain visibility, purpose-built models, and governance.”
– Jeetu Patel, President and Chief Product Officer, Cisco
Cisco originally introduced AgenticOps to redefine how artificial intelligence is applied in networks and address the increasing complexity of modern IT environments. Solutions like Agenttic Workflow and AI Canvas leverage advanced AI and unified network data (including deep network models) to enable faster troubleshooting and secure automation. With the latest updates, Cisco extends these agent capabilities across networking, security, and observability to support IT operations across cloud, on-premises, air-gapped industrial systems, enterprise environments, data centers, and service provider networks.
Operate your network at AI scale with intelligent execution
Cisco is deploying autonomous troubleshooting capabilities across campus, branch, and industrial environments. This enables end-to-end agent investigation to prioritize connectivity and user experience issues, reducing average time to resolution to minutes. Continuous optimization features provide context-aware recommendations designed to prevent performance degradation before users are affected, and trusted validation tools provide risk-aware assessments of network changes based on live topology, configuration, and telemetry data.
New experience metrics unify thousands of network signals into a single actionable view, focusing on user experience metrics such as connection time, capacity, and roaming. Additionally, IT teams can now create production-ready, deterministic automation using agent workflow authoring within Cisco AI Assistant.
In data center environments, AgenticOps enables early detection and intelligent event correlation, providing prescriptive recommendations to optimize network performance. For service providers, new agent capabilities within Crosswork AI accelerate the transition to autonomous networking by quickly and accurately identifying, diagnosing, and resolving complex multivendor problems.
Enhance your security operations with Agentic Intelligence
Within Cisco Security Cloud Control, AgenticOps deploys proactive firewall policy recommendations by analyzing traffic patterns to strengthen zero-trust controls for sensitive applications. The agent’s troubleshooting and optimization capabilities also detect performance-impacting issues such as elephant flows, improving operational efficiency.
Continuous compliance capabilities further strengthen your security operations by automatically assessing firewall configurations, identifying deviations from PCI-DSS, and recommending remediation actions to help your organization remain compliant.
Visibility and control across agent applications
To support observability, Cisco is expanding visibility into the performance of large-scale language models and agent applications. **AI agent monitoring within Splunk Observability Cloud enables teams to track the performance, cost, quality, and behavior of agent workflows, providing deeper insight into how AI-driven operations perform in real-world environments.
With these enhancements, Cisco positions AgenticOps as a fundamental operating model for enterprises seeking to manage AI-scale networks with greater autonomy, governance, and operational efficiency.
