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/degausai/wonda-slidesnpx agentmods add skills/degausai/wonda-slides/slide-generation-systemWrote 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/degausai/wonda-slides/slide-generation-system)<a href="https://agentmods.dev/skills/degausai/wonda-slides/slide-generation-system"><img src="https://agentmods.dev/badge/skills/degausai/wonda-slides/slide-generation-system/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/degausai/wonda-slides/slide-generation-system"><img src="https://agentmods.dev/badge/skills/degausai/wonda-slides/slide-generation-system.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 | $0.00041 | $0.08934 |
| Opus 5 | $0.00020 | $0.04467 |
| Sonnet 5 | $0.00008 | $0.01787 |
| Haiku 4.5 | $0.00004 | $0.00893 |
Grade A, and why
slide-generation-system scanned grade A with 1 finding 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 10d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s https://registry.npmjs.org/<pkg> | jq '{latest: ."dist-tags".latest, maintainers, license}' How it starts
The opening of the file, as written. The whole thing — 809 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: Slide Generation System
Turn a content source (codebase files, Notion notes, Google Docs) into slide-sized (1920×1080) pages that actually look like they came from a specific brand. Output: one HTML file per slide + PDF + PNG.
This is a three-step pipeline:
- Design system: check whether one exists. Use it if it does; extract/create one if it doesn't.
- Content: pull slide copy + assets from a named source (codebase path, Notion page, Google Doc).
- Render: HTML → Playwright → PDF + PNG at the exact target resolution.
If any of the three is skipped or half-done, the output looks generic. All three together is what makes a slide feel branded and on-message.
Who you are
You are a design-systems engineer who refuses to ship generic output. Given a source of branding and a source of content, you produce slides a stranger would immediately recognise as coming from that brand, not a Tailwind default with a color swap, and not a bullet-list dump of the source doc.
Three things must be right, in order:
- Branding fidelity: real tokens AND pattern elements (noise, stripes, dot grids, corner ornaments, gradients, mask fades), not just a color swap.
- Content fidelity: slide text comes from the source, edited for slide format, not invented.
- Export: PDF + PNG at the exact target resolution, saved to a predictable path.
If the render looks like a generic "dark hero with images" or "light card with bullet points", you failed. The pattern stack, typography tokens, and tight content selection are what make it recognisable.
Step 0: Clarify the brief
Do not skip this. Confirm before building:
- Source of branding: repo+app path, or a website URL, or "use the existing design tokens file at X"?
- Source of content: codebase file(s), Notion page URL/ID, Google Doc URL/ID, or a pasted block of text?
- Slide count: one slide or a deck? If a deck: how many, what order, what's the narrative arc?
- Target size: default 1920×1080 (16:9 HD). Alternatives:
1080×1080(Instagram),1200×630(OG),2560×1440(retina),1080×1920(story/reel). - Chrome: logo? Page numbers? Footer URL? Default: NONE. Only add if explicitly asked.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 10d ago First seen · 809 lines · 41 tokens per session scan A 8cf6d35e4148
slide-generation-system is a skill published in the GitHub repository degausai/wonda-slides (2 stars, last pushed 4mo ago), licensed MIT. It adds 41 tokens to every session and 8,934 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
lov-gen-logo
A logo and application-icon design workflow that creates distinct, versioned drafts and keeps brand marks separate from app icons. It also prepares selected assets for their intended uses, such as websites or device screens.
lov-professional-portrait
Turn a single-person photo into a polished professional portrait while preserving the person's identity. Use restrained skin retouching, exposure cleanup, optional hat removal and hairstyle reconstruction, background polish, and an explicit quality gate. Trigger when the user asks for a professional headshot, business…
lov-find-logo
A tool for collecting company or product logos from public web sources such as a website’s icon or social-preview image.
lov-business-card
Generate a professional, editorial-style business card (2:1) as a high-resolution PNG and a self-contained interactive HTML. Collects the user's identity (name, roles, tagline, interests, locations, optional avatar), fills a parameterized design template with a chosen theme, and renders it via headless Chrome. Trigger…
lov-visual-clone
Analyze a reference design image and extract visual DNA — layout, style, color palette, texture, typography, copy tone, spacing, etc. — into a structured, reusable replication prompt that can be applied to new scenarios. Trigger when: user provides a reference image and asks to "extract style", "replicate this"…
lov-riso-portrait
An image-editing workflow that redraws a single-person photo as a recognisable Risograph-style portrait. Risograph is a print style using limited ink colours, visible dots, paper grain, and slight colour misalignment.