modelcontextprotocol / modelcontextprotocol/typescript-sdk
zod v4: "additionalProperties: false" missing
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 13.4k
- Forks
- 2.2k
- Avg merge
- 3d 15h
- Merged PRs (30d)
- 4
Description
Summary
We upgraded zod from v3 to v4. After the upgrade, all input schemas are missing additionalProperties: false in the tools/list response.
With zod@v3 the additionalProperties: false is returned correctly in the inputSchema.
zod: v3 and v4
@modelcontextprotocol/sdk: v1 and v2
Reproduction
Use the following example with zod@v3 and zod@v4:
const server = new McpServer({ name: "test-server", version: "1.0.0" });
server.registerTool(
"test",
{
description: "Breaking input schema",
inputSchema: { testInput: z.string() },
outputSchema: { testOutput: z.string() },
},
async () => ({ content: [], structuredContent: { testOutput: "abc" } })
);
With zod@v4 we have this generated input schema
inputSchema: {
type: "object"
properties: { ... } 1 item
required: [ ... ] 1 item
$schema: "https://json-schema.org/draft/2020-12/schema"
}
But with zod@v3 we have this generated input schema
inputSchema: {
type: "object"
properties: { ... } 1 item
required: [ ... ] 1 item
$schema: "https://json-schema.org/draft/2020-12/schema"
additionalProperties: false
}
Is this expected behavior or a bug?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start at the McpServer.registerTool entry point and trace how its inputSchema is converted for the tools/list response, comparing the zod v3 and v4 reproductions shown here. Confirm the expected behavior with the project’s existing schema-generation tests or add a focused regression test; done means the zod v4 schema includes additionalProperties: false when appropriate.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- typescript
- Domain
- api
- Issue type
- Bug
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Quiet
- Clarity
- Mostly clear
- Newbie friendliness
- 58/100