# Jig

> Jig is a free MCP server for developer tools. Checks JSON against JSON Schema with exact error paths and LLM strict-mode schema checks. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Free
- **Platforms:** Web, API
- **Website:** https://jig.clawhard.com
- **MCP Registry name:** com.clawhard/jig
- **Categories:** [developer tools](https://ai-powered.com/categories/developer-tools)
- **Listing updated:** 2026-11-19

## About

Jig is a validator for the JSON that language models, tool calls and APIs produce. You paste a JSON Schema and a JSON document, and it lists each field that breaks the schema along with its JSON Pointer path, the expected value and the value it received. The retry message it generates can be sent straight back to the model.

Beyond checking data, Jig also checks the schema itself. It reports whether a schema is valid JSON Schema and whether OpenAI strict mode, Claude structured outputs and Gemini accept it, citing each provider rule it finds broken. For each provider it offers a converted copy of the schema that the provider accepts.

The tool runs in the browser without an account. It is also exposed as an MCP server with two tools, so an AI assistant or agent can check model output before writing it to a database or another API. The README and site note that external $ref URLs are not fetched and that some formats are not checked, and provider rules are taken from their documentation.

## Features

- **Path-level validation errors:** Every failing field is reported with its JSON Pointer path, the rule it breaks and what was expected.
- **Multiple draft support:** Handles JSON Schema drafts 2020-12, 2019-09, 07, 06 and 04, including $ref, anyOf, oneOf and if/then/else.
- **Provider strict-mode checks:** Tests a schema against OpenAI strict mode, Claude structured outputs and Gemini, listing each unsupported rule.
- **Converted schema copies:** Produces a version of the schema for each provider that drops or moves the constraints that provider rejects.
- **Model retry messages:** Generates a message describing the validation failures that can be sent back to the model to fix its output.
- **MCP server access:** Exposes jig_validate_json and jig_check_schema tools over Streamable HTTP without authentication.

## Good for

- Check structured output from an LLM before it is saved to a database
- Find why OpenAI or Claude rejects a tool schema sent with strict mode enabled
- Confirm which constraints Gemini drops silently so they can be enforced in code
- Let an AI assistant validate JSON payloads during development
- Feed validation errors back to a model so it can correct its own output

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http jig https://jig.clawhard.com/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://jig.clawhard.com/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "jig": {
      "url": "https://jig.clawhard.com/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "jig": {
      "type": "http",
      "url": "https://jig.clawhard.com/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "jig": {
      "serverUrl": "https://jig.clawhard.com/mcp"
    }
  }
}
```

### Grok Bot (Grok Bot desktop app → Connect apps → Add a custom connector)

```sh
https://jig.clawhard.com/mcp
```

Enter a Name (for example "Jig"), choose Type: Remote HTTPS, paste this as the Server URL and choose Add plugin.

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "jig": {
      "type": "http",
      "url": "https://jig.clawhard.com/mcp"
    }
  }
}
```

## Tools

All 2 tools say they're read-only. From the server's own tool annotations; not verified by us.

- `jig_validate_json` (says read-only): Use when the user asks whether some JSON matches a JSON Schema, or when JSON from a model reply, structured output, tool call, API response or file has to be checked against its schema before it is used or stored (signals: "validate this JSON", "does this match the schema", "which field is wrong", "the model returned bad JSON", "check the tool-call arguments", "why does my parser reject this"). Va
- `jig_check_schema` (says read-only): Use when the user asks whether a JSON Schema is valid, or whether it will work with OpenAI Structured Outputs or function calling with strict: true, Claude structured outputs or strict tool use, or Gemini structured output (response_json_schema), or is writing a tool or function schema, an MCP inputSchema, a response_format, or a schema exported from Zod or Pydantic (signals: "Invalid schema for r

## FAQ

### How do I validate JSON against a JSON Schema?

Paste the schema and the JSON into Jig and press Validate JSON. Each error shows the JSON Pointer path of the bad value, the rule it breaks and what was expected. An AI assistant can do the same through the MCP tool jig_validate_json.

### Why does Claude return a 400 for my tool schema with strict: true?

According to the Jig site, Claude structured outputs reject numeric constraints such as minimum and maximum, string length constraints, minItems above 1, recursive schemas and additionalProperties other than false. Jig counts these and moves unsupported constraints into descriptions.

### Does Gemini enforce maxLength in a response schema?

The Jig site says Gemini's docs state that maxLength is stripped without an error, so the model can break it. Jig flags these constraints so they can be checked in code.

### Does Jig require an account to use?

The site states it runs in the browser with no account, and the MCP server needs no key or login.

### Is Jig free?

Yes. Jig is free to use.

### How do I add Jig to Claude Code?

Run this command in your terminal: claude mcp add --transport http jig https://jig.clawhard.com/mcp

### What are the best alternatives to Jig?

The closest alternatives to Jig are JSON Schema Tools, Json Mender MCP, mcp-lint, ELYREON JSON Schema Payload Validator and LLM Preflight.

## Where this comes from

- Description: written from the project's website and its MCP Registry entry.
- Status and tools: checked automatically by AIPoweredBot on 19 November 2026.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [JSON Schema Tools](https://ai-powered.com/mcp/json-schema-tools): JSON Schema Tools is an open-source MCP server for developer tools.
- [Json Mender MCP](https://ai-powered.com/mcp/json-mender-mcp): Json Mender MCP is an open-source MCP server for developer tools.
- [mcp-lint](https://ai-powered.com/mcp/mcp-lint): mcp-lint is an open-source MCP server for developer tools.
- [ELYREON JSON Schema Payload Validator](https://ai-powered.com/mcp/elyreon-json-schema-payload-validator): ELYREON JSON Schema Payload Validator is a free MCP server for developer tools.
- [LLM Preflight](https://ai-powered.com/mcp/llm-preflight): LLM Preflight is an MCP server for ai models and llm tooling.
- [schema-mcp](https://ai-powered.com/mcp/schema-mcp): schema-mcp is an open-source MCP server for developer tools.
- [ToolSchema Forge](https://ai-powered.com/mcp/toolschema-forge): ToolSchema Forge is a freemium MCP server for ai models and llm tooling.
- [PurRAssist](https://ai-powered.com/mcp/purrassist): PurRAssist is an open-source MCP server for developer tools.

---
Source: https://ai-powered.com/mcp/jig
