Concept·Workflow Infrastructure·Augmentation·Emerging·CON-026
Custom Plugins
Value hypothesis
Designers automate workflows without engineering support, creating custom tooling for tasks not covered by existing plugins.
Efficiency · Innovation
Teams identify repetitive or specialised tasks that would be suitable for automation. The desired functionality is specified in natural language, then produced by an AI coding agent. The designer tests the output, identifies gaps or errors, and re-prompts to refine. Success creates a persistent workflow tool, reusable beyond a single project.
Risks in application
Black Box Rationale
Plugins are built from code the designer cannot read, so when they break or behave unexpectedly they cannot be manually debugged; maintenance depends on re-prompting rather than understanding.
Constraint Blindness
AI-generated plugin code may operate correctly in testing but fail under Figma API updates, edge-case file structures, or team-scale deployment conditions the designer could not anticipate.
Expertise that differentiates
Design System Logic
Specifying plugin functionality addresses real workflow bottlenecks, integrates correctly with design system infrastructure, and is scoped to scale across different use cases within the team's tooling environment.
Technical Feasibility
Assessing whether AI-generated plugin code will behave correctly within API constraints, and recognising implementation approaches likely to be brittle or difficult to maintain across tool versions.
AI Fluency that assures
Creation Diligence
Discipline is necessary to test across the range of actual usage contexts (different file sizes, varied component configurations, target platforms).
Possible Indicators
Automation rate
Ratio of target task steps handled by the plugin without manual intervention
New method adoption
Whether the use case enables automation not previously possible without engineering support
Sources
Fung (2026). Ship It! Vibe Coding Your First Figma Plugin. Into Design Systems AI Conference 2026.
Dane Erick Peralta (2025). From Designer to Developer: How I Built a Figma Plugin Using AI. Medium.
Ford, P. (2026). New work, new words: A glossary for AI [conference talk]. Designing with AI 2026, Rosenfeld.
Oduye, A. (mod.), Crumlish, C., Flowers, E., et al. (2026). From tools to staff: What the next generation of agents means for the future of design [panel]. Designing with AI 2026, Rosenfeld.