“"The question explanations were clear and helped me understand automation concepts instead of simply memorizing answers."”
Emily Walker
Network Engineer
ENAUTO runs 90 minutes and Cisco doesn’t publish an exact question count, though most candidates report somewhere around 55–65 items. You won’t find a lot of pure recall (“what does this acronym stand for”) — the exam leans heavily on scenario format: you’re given a network state, a partial script, or an API response, and asked what’s wrong with it, what the next call should be, or which tool fits the situation. Expect a mix of multiple-choice, multiple-response, and drag-and-drop, with some items asking you to interpret a code snippet rather than write one from scratch.
Cisco moved ENAUTO to version 2.0 on February 3, 2026, and the update reshaped the exam meaningfully — more on that below. If you’re studying from material written before that date, check whether it reflects the current five-domain blueprint before you trust it.
Domain | Weight | Core Focus |
|---|---|---|
1.0 Network Automation Foundation | 10% | YANG data models, NETCONF/RESTCONF fundamentals, building correct XML/JSON payloads |
2.0 Device-Level Network Automation | 25% | Netmiko, ncclient, RESTCONF, and Ansible for single-device automation; Day 0 provisioning; EEM and Guest Shell |
3.0 Controller-Based Network Automation | 30% | Catalyst Center, Catalyst SD-WAN Manager, Meraki Dashboard API, ISE |
4.0 Operations | 20% | pyATS, model-driven telemetry, webhook-based monitoring, SWIM automation |
5.0 AI in Automation | 15% | AI-assisted code development, associated risks, building an MCP server with Python FastMCP |
Controller-Based (30%) and Device-Level (25%) together make up well over half the exam — if your time is limited, that’s where it pays off most to go deep rather than wide.
AI in Automation is not a throwaway domain. At 15%, it’s worth more than the entire foundation domain, and because it’s brand new to v2.0, there’s far less third-party prep content covering it than the older topics. Candidates who skim this section because it “feels soft” tend to lose points they didn’t expect to lose. Know the actual risks of AI-generated automation code — data exposure, intellectual property concerns, validating output before you run it against production — and understand what a Python FastMCP server does and why you’d build one.
Telemetry configuration trips people up more than telemetry theory. It’s one thing to know that model-driven telemetry exists; it’s another to correctly configure a subscription on an IOS XE device across CLI, NETCONF, and RESTCONF and recognize which one a given scenario is using. This shows up repeatedly in the Operations domain and candidates who only studied it conceptually get caught.
Controller terminology has shifted, and the exam uses current naming. DNA Center is now Catalyst Center. vManage is now Catalyst SD-WAN Manager. If your study notes still use the old names, the mapping is easy enough, but don’t let it slow you down mid-exam when a question references the current product name and you pause to translate.
ncclient and RESTCONF look similar on the page and behave differently in practice. Several device-level objectives ask you to construct a solution using one or the other for the same underlying task. Memorizing syntax isn’t enough — you need to recognize which library or protocol a given code fragment is using and why someone chose it.
Because ENAUTO is scenario-heavy, drilling flashcard-style recall won’t get you far. When you’re working through practice questions here, treat each one as a mini case study: before checking the answer, ask yourself which domain it belongs to and why the other options are wrong, not just which one is right. A question that shows you a Netmiko script with a bug is testing whether you understand what the library actually does — if you can only identify the right answer by elimination, go back to the underlying concept before moving on.
Pay particular attention to questions that hand you a code block or an API response and ask you to interpret it. These map directly to the exam’s actual format and are a better use of your remaining study time than questions that only test vocabulary.
Studying Device-Level and Controller-Based automation thoroughly but treating Operations and AI in Automation as an afterthought, despite them combining for 35% of the exam.
Practicing with v1.0-era material that still reflects the old six-domain structure (Network Programmability Foundation, Automate APIs and Protocols, Network Device Programmability, DNA Center, SD-WAN, Meraki) instead of the current five-domain v2.0 blueprint.
Knowing Python concepts in the abstract without having actually run Netmiko, ncclient, and Ansible playbooks against lab devices — the exam expects you to recognize working code, not just discuss automation theory.
Skipping ISE coverage because it feels like a security topic rather than an automation one; it’s folded into the Controller-Based domain and shows up on the exam.
If you’re new to network automation: Budget 8–10 weeks. Spend time with Python basics, YANG/NETCONF/RESTCONF fundamentals, and get comfortable reading API documentation before you touch the controller platforms — trying to learn Catalyst Center automation without first understanding what RESTCONF is doing under the hood will slow you down later.
If you already work with Cisco controllers or scripting day to day: 3–5 weeks is usually . Your gap is more likely to be the AI in Automation domain and the specific telemetry configuration objectives than the fundamentals.
If you’re retaking the exam: Go back to your score report by domain if Cisco provided one, and be honest about whether you actually under-prepared Operations or AI in Automation rather than assuming you just had bad luck on question selection. A week of focused review on your weakest one or two domains is often enough for a retake If you're preparing for another certification, try our PRINCE2 Project Management Practitioner (Version 7) practice test.
ENAUTO has changed substantially more than most CCNP concentration exams. The version 2.0 update, effective February 3, 2026, didn’t just add an objective or two — it restructured the exam from a six-domain, platform-by-platform blueprint (separate domains for DNA Center, SD-WAN, and Meraki) into five domains organized by automation approach instead, and added AI in Automation as an entirely new domain. If you’re using any practice material, course, or study guide published before early 2026, confirm it reflects the current structure rather than the retired one.
Last updated on Oct, 31 2026