The binder that lives in the rack. Documentation generators for small IT teams and MSPs — the documents nobody gets around to writing. Each module takes facts you already have and produces the artefact a client, an auditor or the next tech would ask for.
Live: https://akaushal-nexuscmms.github.io/rackbinder/
Everything runs in the browser. No account, no upload, no database — your
network data never leaves the tab. Drafts persist in localStorage only.
| # | Module | What goes in | What comes out |
|---|---|---|---|
| 1 | VLAN / Subnet Planner | Sites, VLANs with host counts, floors/rooms, subnets already in use | An aligned VLSM plan and an IP addressing-standard document — conventions, per-site VLAN tables, floor/room ranges, retained subnets, growth reserve, static register, change history |
| 2 | Incident Postmortem | Incident facts, pasted notes or a chat log, action items | A blameless postmortem — timeline, time to detect / mitigate / resolve, root cause and contributing factors, what we learned, owned action items, sign-off |
| 3 | Onboarding / Offboarding Builder | planned | |
| 4 | Network Topology Document | planned | |
| 5 | Change Document | planned | |
| 6 | Firewall Rule Change-Log | planned |
Every document exports as Markdown (for a wiki or Git) and as a Word .docx
(for the client), from one shared exporter.
- VLSM, largest-first, so every block is aligned and free space stays contiguous.
- Sizing counts the gateway and the static reserve on top of the host count, so the pool always holds the devices you planned for.
- Multi-site: each site has its own base network; an optional supernet assigns site blocks in one click; site bases are overlap-checked.
- Floors and rooms are contiguous ranges carved from a VLAN's pool, in the order listed — a per-floor DHCP scope, not a separate subnet.
- Brownfield: paste the subnets already live at a site and new VLANs are placed around them. Retained subnets appear in the document.
- Study mode for CCNA revision: every row expands into the worked example (host bits, borrowed bits, block size, magic number, binary split) and a set of practice problems is generated from the plan.
- Messy notes in: lines that start with a time (
10:42,[10:42],8:55 AM,2026-09-05 10:42) become timeline rows; continuation lines merge;Name: …prefixes become the actor. - Key times (fault began / detected / mitigated / resolved) can be filled from timeline rows marked as such; the document reports time-to-detect, time-to-mitigate, time-to-resolve and total duration.
- A blameless-ness lint flags "human error" as a root cause, a person named as the cause, missing owners or due dates, no prevention or detection action, and timestamps that contradict each other.
- Structure follows the Google SRE and PagerDuty postmortem templates.
npm install
npm run dev # http://localhost:3000npm run lint # eslint
node scripts/smoke.ts # ~100 checks over the subnet math, documents, postmortem and DOCX export (Node 22.6+)
npm run build # static export to ./outscripts/ui-review.py drives the dev server with Playwright (Python) and checks
layout, dark mode, phone viewport and the browser console.
Next.js (App Router, static export), TypeScript, Tailwind 4, docx
for Word export. The IPv4 math is hand-written (about forty lines of 32-bit
integer ops) because study mode needs the intermediate steps no library exposes.
Deployed to GitHub Pages by .github/workflows/deploy.yml on every push to
main, after lint, the smoke suite and a build.
Source-available under the Elastic License 2.0: use it, copy it, change it, run it inside your company or for your clients. You may not offer it to others as a hosted or managed service, or circumvent a licence key. Contributions are welcome under the same terms.