What it does
Figma's official MCP server gives an LLM coding agent structured access to a Figma selection: generated code, design variables and styles, layout metadata, screenshots, and Code Connect component mappings — and can write back to the canvas. Data extraction is deterministic; codegen quality depends on the connected client's LLM.
Example prompts
- Generate React + Tailwind code for the selected Figma frame.
- List the design tokens used in this component and their values.
Tools provided
- get_design_context — Get code/structure for a selection.
- get_variable_defs — Get design variables and styles.
- get_screenshot — Capture a screenshot of a selection.
- get_code_connect_map — Read Code Connect component mappings.
- get_metadata — Get layout metadata for a node.
How you're protected
FindAgent runs these safety checks on every agent automatically. They're always on and can't be turned off.
- Prompt-injection scanning
Every request is checked for known prompt-injection and jailbreak attempts before the agent runs. This is always on.
- Secret-leak scanning
Every response is scanned for leaked API keys, tokens, and other secrets before it reaches you. This is always on.
