edit_screens
AI agents use edit_screens to create or update resources in Stitch-MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Stitch-MCP environment.
The tool name 'edit_screens' and the server's explicit mention of 'screen generation/editing' capability indicate this creates or modifies design data reversibly. Changes to UI screens in a design system are not destructive (can be reverted via version control or undo) and do not execute arbitrary code or delete data, making Write the appropriate category.
From the tool's definition Tool name 'edit_screens' indicates modification of screen designs. The description is empty, but context from sibling tools (generate_screen_from_text, generate_variants, create_design_system) and the server's purpose—'screen generation/editing, and design…
Attacks that exploit this kind of access
edit_screens. It is categorised as a Write tool in the Stitch-MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Stitch- MCP server in PolicyLayer and add a rule for edit_screens: 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. Nothing to install.
edit_screens is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the edit_screens 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 edit_screens. 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.
edit_screens is provided by the Stitch- MCP server (samueljayasingh/stitchmcp). 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 →