Medium Risk

sp_design_create

Generate a design system (colors, typography, spacing, rules) from a brand description. Returns a DESIGN.md and an ID for use in generation calls.

How to control sp_design_create ↓

What sp_design_create does on Stitch Pro

AI agents use sp_design_create to create or update resources in Stitch Pro — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Stitch Pro environment.

Medium Risk

Why sp_design_create needs a policy

The tool generates and stores design system configuration files and identifiers. While this is creational and reversible (can be overwritten or deleted), it modifies the project state by adding new design system assets. This does not involve read-only retrieval, destructive deletion, code execution with arbitrary side effects, or financial transactions. It fits Write: creates or modifies data reversibly.

From the tool's definition 'Generate a design system' creates and produces new design artifacts (DESIGN.md and ID) that will be stored and reused. This is a create operation that produces persistent, reversible assets.

Documented attack patterns abuse exactly the kind of access sp_design_create gives an agent:

How to control sp_design_create

PolicyLayer is an MCP gateway — it sits between your AI agents and Stitch Pro, and nothing reaches the server without passing your rules. This is the rule we recommend for sp_design_create:

policy.json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "sp_design_create": {
      "limits": [
        {
          "counter": "sp_design_create_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

sp_design_create stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

  1. Create a free account and register Stitch Pro — nothing to install.
  2. Add this policy — paste it, or build it visually.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
LIMIT THIS TOOL →

Free to start. No card required.

Related tools and policies

Go deeper

Questions about sp_design_create

What does the sp_design_create tool do? +

Generate a design system (colors, typography, spacing, rules) from a brand description. Returns a DESIGN.md and an ID for use in generation calls. It is categorised as a Write tool in the Stitch Pro MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on sp_design_create? +

Register the Stitch Pro MCP server in PolicyLayer and add a rule for sp_design_create: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Stitch Pro. Nothing to install.

What risk level is sp_design_create? +

sp_design_create is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit sp_design_create? +

Yes. Add a rate_limit block to the sp_design_create rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block sp_design_create completely? +

Set action: deny in the PolicyLayer policy for sp_design_create. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides sp_design_create? +

sp_design_create is provided by the Stitch Pro MCP server (luciferdono/stitch-pro-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Stitch Pro tool call.

Start from Stitch Pro, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.

Free to start. No card required.

17 Stitch Pro tools catalogued and risk-classified — across an index of 43,000+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.