figma-context-reader
Read and summarize Figma design context for UI/UX implementation, audit, and design-system work. Use this skill when the user provides a Figma URL, selected Figma node, frame, component, Dev Mode context, variables, library, or screenshot and wants structured facts about hierarchy, layout, tokens, components, assets, variants, interaction notes, or implementation constraints without mutating the Figma file.
How do I install this agent skill?
npx skills add https://github.com/sergekostenchuk/ui-ux-agent-skill-system --skill figma-context-readerIs this agent skill safe to install?
- Gen Agent Trust Hubpass
The figma-context-reader skill is a well-structured, read-only utility for extracting and summarizing design data from Figma. It incorporates privacy policy references and requires evidence labeling for all claims. No security risks were identified.
- Socketpass
No alerts
- Snykpass
Risk: LOW · No issues
What does this agent skill do?
Figma Context Reader
Use this skill to turn Figma context into an evidence-backed design brief. This is a read-only skill; route mutations to figma-canvas-editor or figma-design-system-sync.
Modes
frame-brief: summarize a selected frame or page.component-brief: summarize components, variants, props, and constraints.token-brief: summarize variables, styles, color roles, spacing, radius, typography, and effects.audit-input: collect enough Figma evidence forux-audit-skillordesign-critic-skill.handoff: prepare context for implementation specialists.
Workflow
- Confirm the input type: Figma URL, selected node, file key, screenshot/export, or connected MCP selection.
- Apply
../../shared/privacy-policy.md. - If a
use_figmacall is needed, first invoke the installedfigma:figma-useskill. - Read only the needed nodes, variables, components, styles, and assets. Avoid broad file scans unless the user asks for library/system work.
- Produce a structured context brief with evidence labels. Distinguish direct Figma metadata from visual inference.
- If MCP access is unavailable, ask for an export, screenshot, node metadata, or manual Figma details as fallback.
Output Shape
Use assets/context-brief-template.md.
Evidence Rules
Figma metadata: node names, type, layout fields, variables, styles, component names, asset references.Visual inference: hierarchy, attention, density, likely intent, visual polish.Missing: anything not available from the current Figma context.
Do not claim DOM, runtime behavior, accessibility scan, or implementation evidence from Figma context alone.
Validation And Eval
- Check that the final brief names the evidence source and privacy mode.
- Check that every implementation claim is traceable to Figma metadata, visual inference, or an explicit missing-context note.
- Use
evals/evals.jsonfor forward tests that distinguish connected Figma context from screenshot-only input.
Resources
- Read
$CODEX_SKILLS_DIR/senior-figma-orchestrator/references/figma-mcp-compatibility.mdwhen tool availability is unclear. - Use assets/context-brief-template.md for reports.
How can the creator link this skill?
Add the canonical catalog link to the repository README so users can inspect current installs and available audits. The publishing guide covers the complete discovery path.
<a href="https://skillzs.dev/skills/sergekostenchuk/ui-ux-agent-skill-system/figma-context-reader">View figma-context-reader on skillZs</a>