Compares two screens by name and returns a structured diff showing added elements, removed elements, style changes, and text changes. Useful for tracking design iterations.
AI agents call design_diff to retrieve information from Stitch MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool purely reads and compares two existing screens, returning a structured diff. It has no side effects, does not modify any data, and is used for analysis/tracking purposes only.
From the tool's definition Compares two screens by name and returns a structured diff showing added elements, removed elements, style changes, and text changes
Attacks that exploit this kind of access
Compares two screens by name and returns a structured diff showing added elements, removed elements, style changes, and text changes. Useful for tracking design iterations. It is categorised as a Read tool in the Stitch MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Stitch MCP Server MCP server in PolicyLayer and add a rule for design_diff: 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 MCP Server. Nothing to install.
design_diff is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the design_diff 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.
Set action: deny in the PolicyLayer policy for design_diff. 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.
design_diff is provided by the Stitch MCP Server MCP server (oogleyskr/stitch-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →