skills.rest
automateyournetwork
automateyournetworkCOMMUNITY
Explore all skills by this author →
Website𝕏@john_capobiancoGitHub

comfyui-topology-viz

Generate stylized AI still images of network topologies via a self-hosted ComfyUI instance.

650|182|Updated Feb 19, 2026
Search Tags:#image generation#network topology#Visualization#ComfyUI#stable diffusion#controlnet#FLUX
One-click install
npx skills add https://github.com/automateyournetwork/netclaw --skill comfyui-topology-viz
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: comfyui-topology-viz
Source: https://github.com/automateyournetwork/netclaw/tree/main/workspace/skills/comfyui-topology-viz
Command: npx skills add https://github.com/automateyournetwork/netclaw --skill comfyui-topology-viz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires httpx, mcp, networkx, Pillow.

What problem does it solve?

Network engineers often need a visually striking illustration of a topology for presentations or documentation, but manually drawing stylized diagrams is slow and generic image generators produce pictures that do not match the real network structure.

Core Features & Use Cases

  • Topology-to-image generation: Converts a topology from any of 8 live sources (Cisco Modeling Labs, GNS3, containerlab, EVE-NG, Nautobot, NetBox/Infrahub, IP Fabric, Forward Networks) or a freeform text description into one AI-generated PNG via ComfyUI.
  • Structurally accurate rendering: When Flux and a ControlNet model are installed, a deterministic box/line diagram is rendered first and fed through Canny edge conditioning, so device connections in the image match the real topology; real hostname labels are overlaid deterministically afterward.
  • Distinct failure reporting: Six classified failure kinds (backend unreachable, no usable model, job failed, job already in progress, source unreachable, empty topology) instead of generic errors, with no imposed timeout on GPU generation.
  • Use Case: Ask for a stylized image of your CML lab topology and receive a timestamped PNG plus a sidecar JSON recording the prompt and checkpoint used, saved under workspace/output/comfyui-topology-viz/.

Quick Start

Ask the agent to make a flashy AI-generated image of your CML lab topology and it will return the saved PNG path and the checkpoint used.

Frequently Asked Questions about comfyui-topology-viz

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an AI image of a network topology with ComfyUI?▼

Point COMFYUI_URL at your running ComfyUI instance, then ask for a stylized image of a named topology source such as a CML lab or GNS3 project. The skill retrieves the topology, picks an installed checkpoint, generates one PNG, and saves it with a sidecar JSON under workspace/output/comfyui-topology-viz/.

What topology sources can be rendered as AI-generated images?▼

Eight live sources are supported: Cisco Modeling Labs, GNS3, containerlab, EVE-NG, Nautobot, NetBox/Infrahub, IP Fabric, and Forward Networks. You can also describe a topology in plain language, such as a router connected to a switch, with no live source required.

Does ComfyUI topology generation support video or animation?▼

No, version 1 produces still images only. Video flybys, packet-tracing animations, and stylized test-result cards are explicitly out of scope and reserved for possible follow-on work.

Why does ComfyUI report no usable model found?▼

This happens on a fresh ComfyUI install with no checkpoints downloaded. Install at least one Stable Diffusion 1.5, SDXL, or Flux checkpoint into ComfyUI's models/checkpoints directory using ComfyUI Manager or a manual download, then retry.

Can I run two ComfyUI image generations at the same time?▼

No, the skill enforces a single in-flight job per process. A second request while one is generating is rejected with a distinct already-in-progress message rather than being queued.

How accurate is the generated image to the real network topology?▼

With Flux and a ControlNet installed, a deterministic box/line diagram is rendered and Canny-conditioned, so device connections match the real topology and hostname labels are overlaid deterministically. The plain txt2img fallback has no structural guarantee and may not resemble the actual topology.

Related Skills

Explore More Software Engineering Skills→
dgethingsdgethings

clab-lab-management

Automate ContainerLab network lab lifecycle management via the ContainerLab API.

Community
Advanced
redhat-documentationredhat-documentation

jtbd-compare

Compare feature-based markdown documentation structures with JTBD records.

Official
Intermediate
automateyournetworkautomateyournetwork
627

pyats-topology

Discover and map network topology using CDP, LLDP, ARP, and routing protocols.

Community
Advanced
Top3d-aiTop3d-ai
51

build-world

Convert a reference image into a stylized 3D scene in Blender.

Official
Advanced