mcp-json-yaml-toml 0.5.1

MCP server for querying and modifying JSON, YAML, and TOML configuration files using yq

mcp-json-yaml-toml

A token-efficient, schema-aware MCP server for safely reading and modifying JSON, YAML, and TOML files.

Stop AI coding tools from breaking your data files. No more grep guesswork, hallucinated fields, or invalid configs. This MCP server gives AI assistants a strict, round-trip safe interface for working with structured data.

---

The Problem

AI coding tools often destroy structured data files:

• They grep through huge configs and guess at keys
• They hallucinate fields that never existed
• They use sed and regex that leave files in invalid states
• They break YAML indentation and TOML syntax
• They can't validate changes before writing

The result: Broken deployments, corrupted configs, and manual cleanup work.

The Solution

mcp-json-yaml-toml provides AI assistants with proper tools for structured data:

• Token-efficient queries: Extract exactly what you need without loading entire files
• Schema validation: Enforce correctness using SchemaStore.org or custom schemas
• Safe modifications: Validate before writing; preserve comments and formatting
• Multi-format support: JSON, YAML, and TOML through a unified interface
• Constraint validation: LMQL-powered input validation with partial match support for guided generation
• Local operation: All processing happens locally; one-time initialization may download yq binary for your platform and cache JSON schemas for validation
• Cross-platform: Works on Linux, macOS, and Windows with bundled yq binaries

Compatible with any MCP client: Claude Code CLI, Cursor, Windsurf, VS Code with MCP extensions, and more.

What It Provides

The server provides 7 MCP tools for structured data manipulation:

• data: Get, set, or delete values at specific paths in configuration files
• data_query: Run advanced yq expressions for complex queries and transformations
• data_schema: Validate files against JSON schemas and manage schema catalogs
• data_convert: Convert between JSON, YAML, and TOML formats
• data_merge: Deep merge two configuration files with environment overrides
• constraint_validate: Validate inputs against LMQL constraints with partial match support
• constraint_list: List available constraints for guided generation

See docs/tools.md for detailed API reference and examples.

Key Features

• Powerful querying: Use yq's jq-compatible expressions to extract nested data
• Format conversion: Convert between JSON, YAML, and TOML (with limitations)
• Config merging: Intelligently merge base configs with environment-specific overrides
• YAML optimization: Auto-generate anchors/aliases for duplicate structures (DRY principle)
• Comment preservation: Modifications maintain existing comments and formatting

---

Quick Start

Prerequisites

• Python 3.11 or higher
• An MCP-compatible client (see docs/clients.md for supported clients)

Installation

MCP servers run as external processes and communicate via stdio with your MCP client. This server uses uvx for automatic dependency management.

Claude Code (CLI)

claude mcp add --scope user mcp-json-yaml-toml -- uvx mcp-json-yaml-toml

Other MCP Clients

Most MCP clients use a JSON configuration format. Add this to your client's MCP configuration file:

{
  "mcpServers": {
    "json-yaml-toml": {
      "command": "uvx",
      "args": ["mcp-json-yaml-toml"]
    }
  }
}

See docs/clients.md for detailed setup instructions for specific clients including Cursor, VS Code with Continue, Windsurf, Zed Editor, and other MCP-compatible tools.

Example Usage

Once configured, you can ask your AI assistant to perform operations like:

Reading configuration values:

• "What stages are defined in my GitLab CI file?"
• "Show me the project name from pyproject.toml"
• "Get all dependencies from package.json"

Converting between formats:

• "Convert config.yaml to JSON format"
• "Export this TOML file as YAML"

Note: Conversion TO TOML is not supported. See docs/tools.md for details.

Validation:

• "Validate this YAML file syntax"
• "Check if config.json matches its schema from SchemaStore"

Advanced queries:

• "Extract all job names from .gitlab-ci.yml"
• "Merge base.yaml with production.yaml"
• "Show the structure of config.json without values"

These natural language prompts are translated by your AI assistant into the appropriate MCP tool calls.

---

Available Tools

The server provides 7 tools: data, data_query, data_schema, data_convert, data_merge, constraint_validate, and constraint_list.

See docs/tools.md for parameters, examples, and usage reference.

---

Configuration

See docs/tools.md for environment variables and configuration options.

---

Development

Setup

# Clone repository
git clone https://github.com/bitflight-devops/mcp-json-yaml-toml.git
cd mcp-json-yaml-toml

# Install dependencies (including dev dependencies)
uv sync

Running Tests

# Run all tests with coverage
uv run pytest

# Run specific test file
uv run pytest packages/mcp_json_yaml_toml/tests/test_server.py

# Run with verbose output
uv run pytest -v

Code Quality

# Format code
uv run ruff format

# Lint code
uv run ruff check --fix

# Type check
uv run mypy packages/

# Run all checks
uv run pre-commit run --all-files

---

Project Structure

mcp-json-yaml-toml/
├── packages/mcp_json_yaml_toml/  # Main package
│   ├── server.py                 # MCP server implementation
│   ├── yq_wrapper.py             # yq binary wrapper
│   ├── schemas.py                # JSON Schema management
│   ├── yaml_optimizer.py         # YAML anchor/alias optimization
│   ├── toml_utils.py             # TOML file operations
│   ├── lmql_constraints.py       # LMQL constraint validation
│   ├── config.py                 # Configuration management
│   └── tests/                    # Test suite
├── docs/                         # Documentation
│   ├── tools.md                  # Tool reference
│   ├── clients.md                # Client setup guides
│   ├── module-usage.md           # Module dependencies
│   └── yq-wrapper.md             # yq wrapper usage
├── scripts/                      # Utility scripts
├── fixtures/                     # Test fixtures
├── pyproject.toml                # Project configuration
└── README.md                     # This file

---

Requirements

• Python 3.11 or higher
• Bundled yq binaries (no external installation required)

---

License

MIT License - see LICENSE file for details

---

Contributing

Contributions are welcome! Please ensure all tests pass and code quality checks succeed before submitting a pull request.

Acknowledgments

• Built with FastMCP
• Uses yq for configuration file processing
• Uses LMQL regex derivatives for constraint validation