Skip to content
IsoKron

Every tool carries the applicable hint

The Anthropic connector checklist requires every tool to carry the applicable hint: readOnlyHint: true on read-only tools and destructiveHint: true on tools that modify or delete data; Claude uses them to decide which calls run without confirmation.

Check anthropic-hint · Anthropic connector directory · default severity: blocking · rubric 2.1.0

What fails

The tool sets neither readOnlyHint: true nor destructiveHint at all. These hints decide auto-permissions in Claude: read-only tools run without per-call confirmation and destructive tools always prompt. If destructiveHint is absent the specification's default is true, but the checklist asks for the hint to be stated.

The rule

Every tool must include a title and the applicable hint: readOnlyHint: true for read-only tools, and destructiveHint: true for tools that modify or delete data.

Source: Anthropic connector pre-submission checklist

How to fix it

Set "readOnlyHint": true on tools that only read. On tools that change data set "readOnlyHint": false and an explicit "destructiveHint".

  1. Mark every tool that only reads with readOnlyHint: true (and destructiveHint: false for OpenAI).
  2. On every tool that changes data set readOnlyHint: false and an explicit destructiveHint.
  3. Set openWorldHint too, so the same list passes ChatGPT review.
Before
"annotations": { "openWorldHint": false }
After
"annotations": { "readOnlyHint": false, "destructiveHint": true, "openWorldHint": false }

TypeScript SDK

TypeScript
import { McpServer } from '@modelcontextprotocol/server';
import * as z from 'zod';

const server = new McpServer({ name: 'issues', version: '1.0.0' });

server.registerTool(
  'delete_issue',
  {
    title: 'Delete issue',
    description: 'Delete one issue permanently. Use only when the user explicitly asks to delete an issue.',
    inputSchema: z.strictObject({
      issue_id: z.string().describe('Issue ID, for example ISS-42'),
    }),
    annotations: { readOnlyHint: false, destructiveHint: true, idempotentHint: true, openWorldHint: false },
  },
  async ({ issue_id }) => ({ content: [{ type: 'text', text: 'Done' }] }),
);

Python SDK

Python
from typing import Annotated

from pydantic import Field
from mcp.server.mcpserver import MCPServer
from mcp.types import ToolAnnotations

server = MCPServer("issues")

@server.tool(
    title="Delete issue",
    annotations=ToolAnnotations(readOnlyHint=False, destructiveHint=True, idempotentHint=True, openWorldHint=False),
)
def delete_issue(
    issue_id: Annotated[str, Field(description="Issue ID, for example ISS-42")],
) -> str:
    """Delete one issue permanently. Use only when the user explicitly asks to delete an issue."""
    return "Done"

Run against @modelcontextprotocol/server 2.4.0, mcp 2.3.0 and fastmcp 4.2.2 on 2026-11-18; the checker confirms this check passes on the tool list each snippet produces.

Questions

What happens if I set neither hint?

The specification defaults are readOnlyHint: false and destructiveHint: true, so clients treat the tool as a destructive write and ask before every call. The directory checklist also asks for the hint to be stated, so review sends it back.

Do I need destructiveHint on a read-only tool?

No, the specification ignores it when readOnlyHint is true. OpenAI asks for an explicit destructiveHint: false anyway, so set it.

Are annotations enforced?

No. They are hints clients use for confirmation prompts; the specification tells clients to treat them as untrusted unless the server is trusted.

Further reading

How this check is graded

It counts toward: Anthropic connector directory. See the rubric.

Last verified against Anthropic connector pre-submission checklist on 2026-11-18. The rule above is quoted in the source’s own words. This page as Markdown.

Check your tool list: paste your tools/list, nothing is sent anywhere