USDA FoodData Central nutrition lookup (~400k foods). Search by name (query=cheddar cheese) for matching foods with fdcId, or fetch one food (fdcId=328637) for its full analyzed nutrient profile — energy, protein, fats, carbs, vitamins, minerals with amounts and units, plus ingredients for brande...
AI agents call nutrition.food to retrieve information from Mcp without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and queries nutritional data from a public database (USDA FoodData Central). The operations are strictly read-only: searching foods by name and fetching nutrient profiles. There are no side effects, no data creation or modification, no code execution, no destructive actions, and no financial transactions.
From the tool's definition Tool performs USDA FoodData Central nutrition lookup with 'Search by name' and 'fetch one food' operations. No data modification, deletion, or financial transactions involved. Returns nutrition facts and ingredient information only.
Attacks that exploit this kind of access
USDA FoodData Central nutrition lookup (~400k foods). Search by name (query=cheddar cheese) for matching foods with fdcId, or fetch one food (fdcId=328637) for its full analyzed nutrient profile — energy, protein, fats, carbs, vitamins, minerals with amounts and units, plus ingredients for branded foods. Real analyzed values instead of model-estimated nutrition facts. It is categorised as a Read tool in the Mcp MCP Server, which means it retrieves data without modifying state.
Register the MCP server in PolicyLayer and add a rule for nutrition.food: 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. Nothing to install.
nutrition.food 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 nutrition.food 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 nutrition.food. 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.
nutrition.food is provided by the MCP server (@2sio/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 →