Based on the latest Figma comparison, suggest specific CSS changes to match the design
AI agents call suggest_css_fixes to retrieve information from Mcp Browser Lens without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool reads from the latest Figma comparison result and generates suggestions/recommendations for CSS changes. It does not modify any files, execute code, or apply changes — it is purely an advisory/read operation that surfaces information derived from previously captured data.
From the tool's definition 'suggest specific CSS changes' — the tool only proposes/suggests CSS fixes based on existing comparison data; it does not apply or write any changes
Attacks that exploit this kind of access
Based on the latest Figma comparison, suggest specific CSS changes to match the design. It is categorised as a Read tool in the Mcp Browser Lens MCP Server, which means it retrieves data without modifying state.
Register the Mcp Browser Lens MCP server in PolicyLayer and add a rule for suggest_css_fixes: 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 Mcp Browser Lens. Nothing to install.
suggest_css_fixes 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 suggest_css_fixes 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 suggest_css_fixes. 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.
suggest_css_fixes is provided by the Mcp Browser Lens MCP server (nano-step/mcp-browser-lens). 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 →