Schema | AI Agents Talk to Figma MCP

AI Agents Talk to Figma MCP

Describes the environment variables required to run the server.

Name	Required	Description	Default
`SOCKET_PORT`	No	Port for the WebSocket bridge server. Defaults to 3055. Change to run a second server; the MCP server must be pointed to it with --port.	3055
`FIGMA_MCP_CACHE`	No	Set to 'off' to disable caching of repeated library reads (e.g., get_design_system, get_styles). Caching reduces round trips.	on
`FIGMA_MCP_PROFILE`	No	Tool profile to advertise to the model: 'core', 'standard', or 'full'. Controls the number of tools exposed and per-message token cost.	standard
`FIGMA_ACCESS_TOKEN`	No	Personal access token for the Figma REST API. Required only for the optional comment tools (reading and replying to comments). Not needed for core design tools.
`FIGMA_MCP_SKILLS_DIR`	No	Directory containing custom skill Markdown files. Overrides the default skills directory.	~/.figma-mcp/skills
`FIGMA_MCP_USAGE_FOOTER`	No	Set to 'off' to remove the token usage footer appended to tool results.	on
`FIGMA_MCP_SKILL_AUTOREPAIR`	No	Set to 'off' to disable automatic repair of tool names referenced in skills.	on
`FIGMA_MCP_MAX_RESPONSE_CHARS`	No	Maximum character length for tool responses (~6000 tokens). Adjust to avoid overflowing the context window.	24000

Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.

This server publishes no instructions, or was last inspected before Glama recorded them.

Server capabilities have not been inspected yet.

Functions exposed to the LLM to take actions

Name	Description
No tools

Interactive templates invoked by user choice

Name	Description
No prompts

Contextual data attached and managed by the client

Name	Description
No resources

This server cannot be deployed

ActivityMaintained

ResponsivenessNo issues