Yeah, it's C. Northbound APIs are always about letting external apps talk to controllers like DNAC, not for device-level stuff (that'd be southbound). Pretty sure that's what Cisco means by Intent API here, but open if anyone thinks otherwise.
Free 300-435 Practice Test Questions and Answers (2026) | Cert Empire Practice Questions
Free preview: 20 questions.
Cisco 300-435
What is the correct ncclient method to use to collect the running configuration of a Cisco IOS XE
device that uses NETCONF?get_config method is specifically for fetching configs like running or startup, not operational state. B (get) is tempting but that's for operational data, config needs D. Easy to get those mixed up if you don't check the API docs.get_config is valid for pulling the running config using ncclient with NETCONF. A is used for copying configs, which doesn't match what they're asking here. Pretty sure about this but open to other takes.Definitely D here. get_config is the ncclient method made for pulling device configs like running or startup. The others don't actually retrieve config data directly. Pretty sure that's right, correct me if I'm off.
get_config is the right ncclient method for dumping running config with NETCONF. The others don't match Python client syntax from what I've seen.C/D? SDN is all about overlays (C) and being designed to change/adapt fast (D). Traditional infra is more like A/B, with manual configs box by box. E feels extra, since you can manage SDN with tools even if you aren't deep into software dev. I think C and D fit best here but happy to hear other thoughts.
Which two parameters are mandatory when the Cisco Meraki API is used to create a network?
(Choose two.)type to tell Meraki what kind of network, and organizationId to know where to create it. Timezone defaults if not provided, so not strictly required. Open to correction if Cisco changes this!
Which ncclient method is used to collect the running configuration of a Cisco IOS XE device that uses
NETCONF?get_config() to pull config data, specifically stuff like running-config. get() is more for operational or state info. Pretty sure that's what Cisco is after here, but let me know if I'm missing something.Honestly, I was going to say B because I thought get() grabs all data, but now I'm second guessing since config and operational data use different methods. Anyone seen Cisco ask for running config but accept get() in real exams?
get_config() is the ncclient method for retrieving running or startup config. Option B is a common trap, but get() is used for state/operational data, not actual configuration. Someone correct me if I'm off here.
Cisco SD-WAN deployment must be troubleshooted using vManage APIs. A call to vEdge Hardware
Health API returns the data in the exhibit (only a portion is shown). If this JSON is converted to a
Python dictionary and assigned to the variable “d”, how the status is accessed that is indicated on
line 16?Seriously Cisco, these options are worded to trip you up. It's definitely A because native models are by vendors for their own gear, not cross-platform like OpenConfig. Anyone disagree?
A Python script must be created to deactivate vSmart Policy Cisco SD-WAN vManage Configuration
APIs. The documentation states the URL is as shown in the exhibit for this REST call using POST, and
that “policyId” is a required request parameter. Which line of Python code makes this call, assuming
the variable “s” is a valid Requests session object and the variable “policy-id” is the policyId?I get why D is tempting since a lot of APIs want params in the data body, but the SD-WAN docs show policyId in the URL path for this endpoint. So it's B here. Seen this match up on some hands-on practice too, but open if I've missed something.
My vote is it's B, because vManage API usually expects the policyId in the URL path, not as post data. I saw a similar question on a practice exam and B matched the documentation. Not 100 percent but that's how most SD-WAN REST calls are structured. Anyone disagree?
Guest shell gives the required shell access for zero-touch provisioning on IOS XE. Pretty sure that's what's needed for ZTP to run custom scripts and finish setup.

For these Ansible drag and drop questions, do they always mark "loose" indentation wrong or just if it obviously breaks structure? I get that option D is textbook YAML, but I've seen C work fine in real playbooks. Curious if anyone saw C accepted on recent exams?
Honestly, these drag and drop playbook questions love to trip you up with YAML formatting. From what I've seen, the exam expects perfect indentation like option D-otherwise, Ansible sometimes throws errors even if real-world servers would be fine with C. Pretty sure D is safest, but open to anyone who's tested it live.
The goal is to write a Python script to automatically send a message to an external messaging
application when a rogue AP is detected on the network. The message should include the broadcast
SSID that is in the alert. A function called “send_to_application” is created, and this is the
declaration:
send_to_application(message)
The exhibit also shows the data that is received by the application and stored in the variable
return_val. Which Python code completes the task?

A RESTCONF GET request is sent to a Cisco IOS XE device. The base URL of the request and the
response in XML format are shows in the exhibit. what is the YANG data node that is referenced in
the response?static-routes is actually defined as a container. You can see from the XML in the exhibit that it wraps other nodes, which is what containers do. I'm like 90% on this but correct me if someone thinks otherwise.Pretty sure B is right here, since static-routes is defined as a container in the Cisco YANG model. The exhibit's XML shows that structure too. Seen similar layout in practice labs. Could be wrong though if the question meant something more specific.
Refer to the exhibit. The lab group consists of four Cisco IOS XE routers named pod-11,pod-12, and
pod-22. What is the result of running the Ansible playbook to reset the lab?I usually double-check the official Cisco docs for playbook behavior like this, plus trying it in an actual lab. I think it's C since the playbook targets Loopback1 on pod-11 and pod-12, so only those interfaces get removed. Not fully sure though if there's tricky task templating. Anyone reference a practice exam yet?
My vote is C fits here. The Ansible playbook targets Loopback1 and specifically names pod-11 and pod-12, so it removes that interface from just those routers. Not totally sure unless I saw the full playbook but from past labs I'm pretty confident. If anyone's seen different behavior let me know!

PUT to the right endpoint, not just a GET. Looks like the exam drags some folks with similar endpoints for reading vs updating configs. Pretty sure C is correct but open to someone pointing out if there's a catch.
Refer to the exhibit. Interfaces named Loopback0, Loopback1, and Loopback2 must be created and
enabled on a Cisco IOS XE target device in the lab group. Which loop must be added to the end of the
Ansible “create int” task?Had something like this in a mock, and it's A. with_items iterates through the interface list variable in Ansible playbooks, so matches the intent here. The others aren’t valid loop keywords for this use. Anyone disagree?
A tbh. Only with_items handles a list like intlist in Ansible, that's the classic syntax. C (with_list) looks tempting but isn't actually valid in standard playbooks from what I've seen, it's kind of a trap answer. Could be wrong if Ansible changed something recently though, anyone seen otherwise?