AI agents call filter_recipes_by_ingredient to retrieve information from Paprika without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and filters existing recipe data based on ingredient criteria. It has no side effects, does not modify data, and does not execute code or perform destructive operations. It is a straightforward data retrieval function analogous to a search or list operation, placing it firmly in the Read category with low severity.
From the tool's definition Tool name 'filter_recipes_by_ingredient' and description 'Filter recipes by specific ingredients' indicate a query/retrieval operation.
Attacks that exploit this kind of access
Filter recipes by specific ingredients. It is categorised as a Read tool in the Paprika MCP Server, which means it retrieves data without modifying state.
Register the Paprika MCP server in PolicyLayer and add a rule for filter_recipes_by_ingredient: 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 Paprika. Nothing to install.
filter_recipes_by_ingredient 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 filter_recipes_by_ingredient 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 filter_recipes_by_ingredient. 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.
filter_recipes_by_ingredient is provided by the Paprika MCP server (sandordaroczi/paprika-mcp-python-server). 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 →