create_recipe · Paprika MCP: Risk & Policy

WHAT IT DOES

What create_recipe does on Paprika

AI agents use create_recipe to create or update resources in Paprika — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Paprika environment.

WHY IT NEEDS A POLICY

Why create_recipe needs a policy

This tool creates new recipe data in the Paprika system, which is a reversible Write operation (recipes can be deleted or modified). While it incurs a small financial cost (~$0.002 per auto-generated image), the primary function is data creation, not financial commitment—the cost is a secondary effect of the optional image generation feature.

From the tool's definition Tool description states 'Create a new recipe in Paprika' and mentions auto-generating images via external API (Replicate). The name 'create_recipe' and description explicitly indicate data creation with reversible side effects.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

create_recipe Rate-limited

create_recipe stays usable, but capped — an agent stuck in a loop can't make hundreds of changes a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "create_recipe": {
      "limits": [
        {
          "counter": "create_recipe_rate",
          "window": "minute",
          "max": 30,
          "scope": "grant"
        }
      ]
    }
  }
}

LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Paprika, so you can run agents against it safely.

FAQ

Questions about create_recipe

What does the create_recipe tool do? +

Create a new recipe in Paprika. Auto-generates professional AI food photography (~$0.002 per image using Replicate). Set generate_image=false to disable. It is categorised as a Write tool in the Paprika MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

How do I enforce a policy on create_recipe? +

Register the Paprika MCP server in PolicyLayer and add a rule for create_recipe: 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.

What risk level is create_recipe? +

create_recipe is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit create_recipe? +

Yes. Add a rate_limit block to the create_recipe 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.

How do I block create_recipe completely? +

Set action: deny in the PolicyLayer policy for create_recipe. 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.

What MCP server provides create_recipe? +

create_recipe 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.