AI agents call search_cocktail to retrieve information from Cocktail without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool queries cocktail recipe data from TheCocktailDB API and returns information (ingredients, measurements, instructions, image URL) without creating, modifying, deleting, or executing any operations. It is a straightforward read operation with no blast radius if misused by an AI agent.
From the tool's definition Tool name 'search_cocktail' and description 'Search for cocktail recipes by name. Returns ingredients, measurements, instructions, and image URL.' indicate data retrieval only with no side effects.
Attacks that exploit this kind of access
Search for cocktail recipes by name. Returns ingredients, measurements, instructions, and image URL. It is categorised as a Read tool in the Cocktail MCP Server, which means it retrieves data without modifying state.
Register the Cocktail MCP server in PolicyLayer and add a rule for search_cocktail: 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 Cocktail. Nothing to install.
search_cocktail 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 search_cocktail 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 search_cocktail. 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.
search_cocktail is provided by the Cocktail MCP server (yusuke-shibata23/cocktail-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 →