Suggests architecture patterns for specific use cases based on industry best practices. Provides implementation examples and considerations for the recommended patterns. Includes diagrams and explanations of pattern benefits and tradeoffs. Uses the configured Vertex AI model (${modelIdPlaceholder...
AI agents call architecture_pattern_recommendation to retrieve information from Google AI Search MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This is an informational/advisory tool that retrieves and synthesizes existing architecture knowledge without creating, modifying, deleting data or executing code. The tool queries external sources (Google Search via Vertex AI) to generate recommendations, which is a read operation. The output is guidance and documentation, not state changes in any system.
From the tool's definition Tool 'suggests' and 'provides' architecture patterns, examples, diagrams, and explanations based on existing knowledge and web search results. No modification, deletion, code execution, financial transaction, or irreversible action is described or implied.
Attacks that exploit this kind of access
Suggests architecture patterns for specific use cases based on industry best practices. Provides implementation examples and considerations for the recommended patterns. Includes diagrams and explanations of pattern benefits and tradeoffs. Uses the configured Vertex AI model (${modelIdPlaceholder}) with Google Search. Requires. It is categorised as a Read tool in the Google AI Search MCP MCP Server, which means it retrieves data without modifying state.
Register the Google AI Search MCP server in PolicyLayer and add a rule for architecture_pattern_recommendation: 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 Google AI Search MCP. Nothing to install.
architecture_pattern_recommendation 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 architecture_pattern_recommendation 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 architecture_pattern_recommendation. 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.
architecture_pattern_recommendation is provided by the Google AI Search MCP server (shariqriazz/google-ai-search-mcp). 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 →