A policy change is a two-command experiment: capture the state, apply, capture again. Build the two or three routers involved with their current configs, run show ip bgp or show ip route and the neighbour summary, apply the route-map, prefix-list, ACL or interface cost, and run the same commands. The difference is the change's whole effect. In ChatGPT, Claude or Meta Muse with NetPilot connected, the assistant builds the routers, keeps the exact commands and quotes both outputs, so the same commands verify the change on the production devices during the window.
Build only the routers the policy touches
The router you edit plus one neighbour on each side is enough to see the effect. Give the assistant their current configs and it deploys them on real images in your lab VM.
Run the before commands
One command for the state the policy acts on (the BGP table, the route, the interface cost, the ACL counters) and one for the neighbour. The assistant quotes the output.
Apply the exact lines
Paste the route-map, prefix-list, access-list or interface line as you will push it. For a BGP policy, follow with a soft clear so the policy is re-evaluated without resetting the session.
Run the same commands again
Read the two outputs next to each other. The lines that differ are the change. Anything else that moved is a side effect you now know about before the window.
Say this to the assistant
Build R1, R2 and R3 on Cisco IOL in a triangle running OSPF area 0, with a loopback 10.3.0.1/24 on R3. On R1 run show ip ospf interface brief and show ip route 10.3.0.0. Then set ip ospf cost 50 on R1 GigabitEthernet0/2, run the same two commands and show me both outputs.
R1 advertises two prefixes to AS 65002 and the change is an outbound route-map that keeps 10.99.0.0/16 inside. The before command is the advertised-routes list for that neighbour. The change is the prefix-list, the route-map, the neighbour line and a soft clear outbound so the policy runs without resetting the session.
ip prefix-list NO-LAB seq 5 deny 10.99.0.0/16 ip prefix-list NO-LAB seq 10 permit 0.0.0.0/0 le 32 route-map TO-AS65002 permit 10 match ip address prefix-list NO-LAB router bgp 65001 neighbor 10.0.12.2 route-map TO-AS65002 out end clear ip bgp 10.0.12.2 soft out
R1# show ip bgp neighbors 10.0.12.2 advertised-routes
Network Next Hop Metric LocPrf Weight Path
*> 10.1.0.0/24 0.0.0.0 0 32768 i
*> 10.99.0.0/16 0.0.0.0 0 32768 i
Total number of prefixes 2R1# show ip bgp neighbors 10.0.12.2 advertised-routes
Network Next Hop Metric LocPrf Weight Path
*> 10.1.0.0/24 0.0.0.0 0 32768 i
Total number of prefixes 1The ACL denies telnet inbound on GigabitEthernet0/1 and permits everything else. The before output proves the list exists and the interface has no inbound ACL. After the change, the interface shows the list and the match counters on each line say what it is catching and what it is letting through once a host behind it sends traffic.
ip access-list extended BLOCK-TELNET deny tcp any any eq 23 permit ip any any interface GigabitEthernet0/1 ip access-group BLOCK-TELNET in
R1# show ip access-lists BLOCK-TELNET
Extended IP access list BLOCK-TELNET
10 deny tcp any any eq telnet
20 permit ip any any
R1# show ip interface GigabitEthernet0/1 | include access list
Inbound access list is not setR1# show ip access-lists BLOCK-TELNET
Extended IP access list BLOCK-TELNET
10 deny tcp any any eq telnet (6 matches)
20 permit ip any any (142 matches)
R1# show ip interface GigabitEthernet0/1 | include access list
Inbound access list is BLOCK-TELNETR1 reaches 10.3.0.0/24 directly over GigabitEthernet0/2 at metric 2. Raising the cost on that interface to 50 should move the route through R2 instead. The before output shows the cost column and the current next hop. The after output shows the new cost and the route re-installed via GigabitEthernet0/1 at metric 3, a few seconds old.
interface GigabitEthernet0/2 ip ospf cost 50
R1# show ip ospf interface brief Interface PID Area IP Address/Mask Cost State Nbrs F/C Gi0/1 1 0 10.0.12.1/30 1 P2P 1/1 Gi0/2 1 0 10.0.13.1/30 1 P2P 1/1 R1# show ip route 10.3.0.0 Routing entry for 10.3.0.0/24 Known via "ospf 1", distance 110, metric 2, type intra area * 10.0.13.2, from 10.255.0.3, 00:12:40 ago, via GigabitEthernet0/2
R1# show ip ospf interface brief Interface PID Area IP Address/Mask Cost State Nbrs F/C Gi0/1 1 0 10.0.12.1/30 1 P2P 1/1 Gi0/2 1 0 10.0.13.1/30 50 P2P 1/1 R1# show ip route 10.3.0.0 Routing entry for 10.3.0.0/24 Known via "ospf 1", distance 110, metric 3, type intra area * 10.0.12.2, from 10.255.0.3, 00:00:09 ago, via GigabitEthernet0/1
The assistant runs the commands through NetPilot and quotes the output in the chat. Every device in the lab is also a normal SSH target, so you can open the router yourself and read show ip route ospf by hand, as in this capture from the NetPilot web app.
The same before-and-after method for any change, with the per-OS command table.
Mirror lab, pre and post snapshot, the diff a change board signs off on.
What a route-map changes in the best-path decision, with the show ip bgp columns to read.
Build and verify a multi-router OSPF topology, including the interface cost.
How to connect, what the tools do, which plan is needed, lab credits and the idle timeout.
What each plan includes, with lab credits and idle timeouts.
Create a NetPilot account, pick Pro or above, and connect NetPilot in ChatGPT, Claude or Meta Muse.