SamuraiBuddha/mcp-time-precision

MCP Time Precision Server

🕐 A high-precision time MCP (Model Context Protocol) server providing microsecond-accurate timestamps for multi-instance Claude coordination.

Features

• Microsecond Precision: Timestamps accurate to 6 decimal places (YYYY-MM-DDTHH:MM:SS.ffffff)
• Epoch Microseconds: Unix epoch time in microseconds for precise ordering
• Instance Identification: Unique instance IDs for multi-Claude deployments
• Collision Detection: Built-in support for detecting simultaneous operations
• Timezone Support: Full IANA timezone database support
• MAGI Integration: Pre-configured instance names for the Three Wise Men setup

Why Microsecond Precision?

When running multiple Claude Desktop instances (e.g., on Melchior, Balthasar, and Caspar machines) that share a common Memory(Ref) knowledge graph:

• Second precision: High collision probability with 3+ instances
• Millisecond precision: 1/1000 collision probability per second
• Microsecond precision: 1/1,000,000 collision probability per second

Installation

Via pip

pip install mcp-time-precision

Via uv (recommended)

uvx mcp-time-precision

From source

git clone https://github.com/SamuraiBuddha/mcp-time-precision.git
cd mcp-time-precision
pip install -e .

Configuration

Claude Desktop

Add to your Claude Desktop configuration:

{
  "mcpServers": {
    "time-precision": {
      "command": "python",
      "args": ["-m", "mcp_time_precision", "--instance-id", "Melchior-001"]
    }
  }
}

Multi-Instance Setup (MAGI)

For the Three Wise Men configuration:

Melchior (CAD/3D Machine):

{
  "mcpServers": {
    "time-precision": {
      "command": "python",
      "args": ["-m", "mcp_time_precision", "--instance-id", "Melchior-001"]
    }
  }
}

Balthasar (AI/LLM Machine):

{
  "mcpServers": {
    "time-precision": {
      "command": "python",
      "args": ["-m", "mcp_time_precision", "--instance-id", "Balthasar-001"]
    }
  }
}

Caspar (Code/Data Machine):

{
  "mcpServers": {
    "time-precision": {
      "command": "python",
      "args": ["-m", "mcp_time_precision", "--instance-id", "Caspar-001"]
    }
  }
}

Available Tools

get_precise_time

Get current time with microsecond precision in ISO format.

Parameters:

• timezone (optional): IANA timezone name (default: system timezone)

Returns:

{
  "timestamp": "2025-06-16T11:30:45.123456",
  "timezone": "America/New_York",
  "instance_id": "Melchior-001"
}

get_epoch_micros

Get current Unix epoch time in microseconds.

Returns:

{
  "epoch_micros": 1718553045123456,
  "instance_id": "Melchior-001"
}

get_instance_info

Get information about this MCP instance.

Returns:

{
  "instance_id": "Melchior-001",
  "start_time": "2025-06-16T11:00:00.000000",
  "uptime_seconds": 1845.123456,
  "system_info": {
    "hostname": "melchior-workstation",
    "platform": "Windows-11"
  }
}

convert_time_precision

Convert between different time precisions.

Parameters:

• time (required): Time string or epoch value
• input_format: "iso", "epoch_seconds", "epoch_millis", "epoch_micros"
• output_format: "iso", "epoch_seconds", "epoch_millis", "epoch_micros"
• timezone (optional): For ISO format conversions

Usage Examples

Memory Breadcrumb Pattern

# Get microsecond timestamp for breadcrumb
result = await mcp.call_tool("get_precise_time")
timestamp = result["timestamp"]
instance = result["instance_id"]

# Create breadcrumb: [2025-06-16T11:30:45.123456][Melchior-001][OPERATION][STATUS]
breadcrumb = f"[{timestamp}][{instance}][DOCKER_START][SUCCESS][Started n8n container]"

Collision Detection

# Get epoch microseconds for precise ordering
result = await mcp.call_tool("get_epoch_micros")
epoch_micros = result["epoch_micros"]

# Use for distributed consensus or operation ordering
operation_id = f"{epoch_micros}-{result['instance_id']}"

Development

Running Tests

pytest tests/

Debug Mode

python -m mcp_time_precision --debug --instance-id "Test-001"

MCP Inspector

npx @modelcontextprotocol/inspector python -m mcp_time_precision

Architecture

• Built on MCP SDK for standardized protocol support
• Uses Python's time.time() for microsecond precision
• Thread-safe implementation for concurrent access
• Minimal dependencies for reliability

Contributing

Contributions welcome! Especially interested in:

• Performance optimizations for high-frequency calls
• Additional time manipulation tools
• Integration with distributed systems
• Alternative precision options (nanoseconds?)

License

MIT License - see LICENSE file

Acknowledgments

• Inspired by the need for precise multi-instance coordination in CORTEX
• Named after the MAGI supercomputer from Evangelion
• Built for the BIM/AI integration future

"Time is an illusion. Microsecond time, slightly less so." - With apologies to Douglas Adams