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/PMDevSolutions/Aureliusnpx agentmods add skills/pmdevsolutions/aurelius/canva-token-inferenceWrote 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/pmdevsolutions/aurelius/canva-token-inference)<a href="https://agentmods.dev/skills/pmdevsolutions/aurelius/canva-token-inference"><img src="https://agentmods.dev/badge/skills/pmdevsolutions/aurelius/canva-token-inference/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/pmdevsolutions/aurelius/canva-token-inference"><img src="https://agentmods.dev/badge/skills/pmdevsolutions/aurelius/canva-token-inference.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 | $0.00000 | $0.03077 |
| Opus 5.5 | $0.00000 | $0.01231 |
| Sonnet 5.5 | $0.00000 | $0.00615 |
| Haiku 4.5 | $0.00000 | $0.00308 |
Grade A, and why
canva-token-inference 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 — 325 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Canva Token Inference — AI-Powered Token Extraction
Purpose
Extract design tokens from Canva design screenshots using Claude's vision capabilities. Unlike Figma's programmatic token extraction, Canva doesn't expose design data through APIs. This skill bridges that gap by analyzing screenshots to infer colors, typography, spacing, and effects, then presenting them with confidence scores for user confirmation before writing the lockfile.
The output is identical to design-token-lock — a design-tokens.lock.json that downstream phases consume with no awareness of the source.
When to Use
- Phase 2 of the
/build-from-canvapipeline (aftercanva-intake) - Phase 2 of the
/build-from-screenshotpipeline (afterscreenshot-intake) - Any time you need to extract design tokens from Canva screenshots
- Any time you need to extract design tokens from screenshots (any source)
- When regenerating Tailwind config from Canva screenshots
Inputs
- Required:
.claude/plans/build-spec.jsonwith"source": "canva"or"source": "screenshot"and screenshot paths- For canva: reads
canva.exportedScreenshots[] - For screenshot: reads
screenshot.capturedScreenshots[]
- For canva: reads
- Optional: User-provided brand guidelines, style guide, or color palette
Process
Step 1: Gather Screenshots for Analysis
Read build-spec.json and collect all exported screenshots:
1. Determine source type from build-spec.json.source
2. If "canva": read canva.exportedScreenshots[]
3. If "screenshot": read screenshot.capturedScreenshots[]
4. Verify all screenshot files exist
5. If missing:
- canva: re-export via Canva MCP
- screenshot URL: re-capture via Chrome DevTools/Playwright MCP
- screenshot files: ask user to re-provide
Step 2: AI Vision Token Extraction
Analyze each screenshot with Claude vision. Extract tokens in stages:
Pass 1 — Colors:
Analyze the screenshots and extract ALL distinct colors used in the design.
Group them as:
1. Brand/Primary colors (dominant, used for CTAs and key UI)
2. Secondary/Accent colors
3. Neutral/Gray scale (backgrounds, borders, muted text)
4. Semantic colors (success green, error red, warning amber, info blue)
5. Background colors (page, card, section backgrounds)
6. Text colors (headings, body, muted, links)
For each color, provide:
- Exact hex value (best estimate from the screenshot)
- Suggested semantic name
- Where it appears in the design
- Confidence: high (clear, solid color) / medium (gradient or subtle) / low (ambiguous)
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 First seen · 325 lines · 0 tokens per session scan A 71afd186eb8f
canva-token-inference is a skill published in the GitHub repository PMDevSolutions/Aurelius (9 stars, last pushed yesterday), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,077 tokens. 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 2026-09-04.
Other skills, from other repositories
mk:figma
Read-first Figma gateway via Figma MCP: analyze designs, implement Figma-to-code for 1–3 screens, extract design tokens, produce a Figma Evidence Packet for planning handoff, screenshot fallback. Advanced operations (Code Connect, canvas writes, design-system/library patterns) are gated references loaded only on…
generate-project-design-system
A draft generator that studies a project's code and Figma design data to prepare a design-system skill. A design system is a shared set of colors, sizes, components, and usage rules for keeping interfaces consistent.
figma-codegen
Generate framework-aware code from a Figma design. Reads the project's stack profile and emits code matching the existing framework (React/Vue/Svelte/Next/etc.) and styling (Tailwind/CSS/CSS-in-JS), reusing existing components and design tokens instead of regenerating from scratch. Triggers whenever the user wants a…
figma-build
Build a Figma design from code or a description — the reverse of figma-codegen. Reuses the connected file's existing design system (components, variables, styles) instead of drawing primitives with hardcoded values. Triggers whenever the user wants something created or updated IN Figma from code or a spec — e.g.…
refero-design
Primary/default skill for UI design, product design, web design, landing pages, dashboards, product screens, redesigns, visual polish, frontend/CSS styling, design systems, components, responsive design, typography, color, spacing, motion, icons, accessibility, copywriting, conversion, and anti-AI-slop work. Use this…
producao-no-figma
Orientação para organização e produção em Figma: componentes, auto-layout e sistemas de design.