Skip to main content
Register a prompt when you want clients to fetch reusable prompt material from your server.

​
Choose your input model

Use arguments for named string inputs. Use argumentsSchema when the handler should receive parsed and validated values.

​
Register a prompt with arguments

When you define prompt arguments inline, Redop infers their shape in the handler. In this example, args.code is required and args.language is optional. 
app.prompt("code_review", {
  description: "Review a code snippet",
  arguments: [
    { name: "code", required: true },
    { name: "language" },
  ],
  handler: ({ arguments: args }) => [
    {
      role: "user",
      content: {
        type: "text",
        text: `Review this ${args.language ?? "code"}:\n${args.code}`,
      },
    },
  ],
});

​
Register a prompt with argumentsSchema

Use argumentsSchema when the prompt arguments need parsing or validation. 
import { z } from "zod";

app.prompt("summarise", {
  description: "Summarise a topic",
  argumentsSchema: z.object({
    limit: z.coerce.number().int().min(1),
    topic: z.string().min(1),
  }),
  handler: ({ arguments: args }) => [
    {
      role: "user",
      content: {
        type: "text",
        text: `Summarise ${args.topic} in ${args.limit} sentences.`,
      },
    },
  ],
});

​
Use both fields carefully

You can define both arguments and argumentsSchema on the same prompt. When you do:
  • arguments controls MCP prompt metadata
  • arguments controls the early missing-required-argument check
  • argumentsSchema controls parsing and handler typing
Redop does not currently reject mismatches between them. Keep them aligned if you use both.

​
When prompts fit best

  • reusable prompt building blocks
  • prompt material that should stay alongside your server logic
  • flows where the client wants prompt metadata and arguments, not a tool call