Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/Morningstar202604/awesome-skillkitnpx agentmods add skills/morningstar202604/awesome-skillkit/neural-net-drawWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/morningstar202604/awesome-skillkit/neural-net-draw)<a href="https://agentmods.dev/skills/morningstar202604/awesome-skillkit/neural-net-draw"><img src="https://agentmods.dev/badge/skills/morningstar202604/awesome-skillkit/neural-net-draw/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/morningstar202604/awesome-skillkit/neural-net-draw"><img src="https://agentmods.dev/badge/skills/morningstar202604/awesome-skillkit/neural-net-draw.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00090 | $0.01798 |
| Opus 5.5 | $0.00036 | $0.00719 |
| Sonnet 5.5 | $0.00018 | $0.00360 |
| Haiku 4.5 | $0.00009 | $0.00180 |
Grade A, and why
neural-net-draw scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Neural Net Drawer (SOTA)
Generate compile-ready neural-network structure TikZ from a layer spec. Supports two drawing modes:
- per-neuron (v1-compatible): draws one dot per neuron per layer (at most 10 visible dots per layer when very wide; layer height capped at 3cm);
- typed-blocks (PlotNeuralNet style): draws colored
nnblocks by layer type (conv/pool/residual/linear/attention/...), closer to a real architecture.
Honest disclosure:
compilable: trueis an assertion of static syntactic self-consistency — the script doesn't run pdflatex; it only mathematically guarantees TikZ self-consistency. Whether compile passes depends on whether the preamble includes pgfplots + tikz.
Input Checklist
| Input | Required | Notes |
|---|---|---|
| layer spec | yes | --layers: comma-separated. Legacy pure-width "784,512,10" (default fully-connected) or typed "784:input,64:conv,64:pool,128:linear,10:output". Legal types: input,conv,pool,residual,linear,fc,attention,dropout,output |
| figure caption label | no | --label "CNN+MLP" (default Model; written into the generated file's first-line comment) |
| activation function name | no | --activation relu (default relu; only written to output JSON metadata, doesn't change the drawing) |
| drawing style | no | --style auto/per-neuron/typed-blocks (default auto: typed -> typed-blocks, otherwise per-neuron) |
| output path | no | --output net.tex; by default always writes to disk at /tmp/nn_model.tex |
When missing, ask everything at once: "Please provide: ① layer spec --layers (may include types) ② caption --label (optional) ③ whether to specify --output (otherwise writes to /tmp/nn_model.tex)."
Pre-flight Checks
python3 --version # expect >= 3.8, else error and STOP
test -f scripts/neural_net_draw.py && echo OK # expect OK printed, else script missing STOP
If the artifact needs a compile preview: which pdflatex, and the preamble needs \usepackage{pgfplots} + \usepackage{tikz}; without a LaTeX environment, deliver only the .tex source and don't claim compilation was verified.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · -6 tokens per session ca2bebdf6279
- 1mo ago Changed · +9 lines · -13 tokens per session 135ea49d8911
- 1mo ago First seen · 93 lines · 109 tokens per session scan A 12629d4e62cb
neural-net-draw is a skill published in the GitHub repository Morningstar202604/awesome-skillkit (0 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 90 tokens to every session and 1,798 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-19.
Other skills, from other repositories
schematics
Creates or refines publication-quality technical diagrams with the native generateimage tool (Nano Banana Pro through Ace or a Gemini key, GPT Image 2 through an OpenAI key), method and architecture overviews, pipelines, CONSORT and PRISMA flows, experimental workflows, biological pathways, circuits and conceptual…
latex-posters
Create professional research posters in LaTeX using beamerposter, tikzposter, or baposter. Support for conference presentations, academic posters, and scientific communication. Includes layout design, color schemes, multi-column formats, figure integration, and poster-specific best practices for visual communication.
paperbanana-diagram
Author publication-quality diagram specifications (method.txt files) from paper prose and draw.io sketches, following federal data visualization standards and Tufte design principles. Use when creating, modifying, or regenerating conceptual diagrams, architecture figures, or visual explanations for academic papers or…
beautiful-mermaid
Build publication-grade Mermaid diagrams with consistent styling and SVG-first export for manuscript assets.
drug-discovery
Drug discovery: ChEMBL search, drug-likeness, interactions.
archify
Validated interactive HTML diagrams, upstream-maintained.