Routing Lab as Code
SHIPS WHENA public repo where one command builds the fabric and a second asserts converged state: adjacencies up, routes present, paths taking the intended AS hops.
The problem
My network experience is real but it is proxy-shaped: VPN internals, TLS, firewall policy, reverse proxies. None of it is routing protocol evidence, and a network role that screens on BGP and OSPF has no way to see the rest of it. This exists to close that specific gap.
The scope
A multi-AS topology on containerlab with FRR nodes, where the addressing plan, the peerings and the protocol config are generated from a single declarative file instead of hand-written per device.
The part that matters is the validation, not the topology. There are a great many public spine-leaf repos and almost none of them prove the fabric converged. The plan is config pre-checks with Batfish and post-deploy state assertions in pytest, so a broken change fails in CI rather than in a screenshot.
Scope and limits
Every project page on this site carries this section. If a tool is not ready for something, the honest place to say so is next to the claim, not three pages into a README.
- No repository exists yet. This entry describes committed scope, not shipped work.
- All software routing. No vendor hardware and no claim of vendor platform experience.
Building something in this territory, or hiring for it?