MCP Service / AI Programming

Added in version 7.6.0

The MCP Service is a feature of DataFlux Func designed to support AI Agent tool invocation.

Users can allow AI Agents to participate in writing functions within DataFlux Func via the MCP Service.

1. Access Steps

Follow the steps below to connect an MCP client / AI Agent to DataFlux Func's MCP Coding service.

1.1 Create Access Token

Accessing DataFlux Func's MCP Coding service requires creating an Access Token and enabling the "MCP Programming" feature.

You can also specify the functions to be exposed in the rules (e.g., by Script Set ID).

1.2 Permission Rules

Permission rules can be manually edited. They are matched sequentially from top to bottom, and the first matching rule determines the permissions.

Examples are as follows:

Rule (Exact ID Match)	Matching Method
category:math:r	Functions where category is math
scriptSet:mytest	Functions belonging to Script Set mytest
script:mytest__demo	Functions belonging to Script mytest__demo
func:mytest__demo.func	Function with ID mytest__demo.func

Rule (Wildcard ID)	Matching Method
scriptSet:myte*	Functions matching Script Set ID wildcard myte*
script:mytest__de*	Functions matching Script ID wildcard mytest__de*
func:mytest__demo.fu*	Functions matching Function ID wildcard mytest__demo.fu*

Rule (Deny)	Matching Method
!category:math:r	Deny functions where category is math

1.3 Configure MCP Client

DataFlux Func's MCP service adheres to the MCP specification

It is recommended to use MCP Inspector for debugging before actual integration.

Add the following configuration to your MCP client / AI Agent:

Configuration Item	Value
URL Address	{DataFlux Func Access URL}/mcp/coding
Request Headers	Authorization: Bearer atk-xxxxx:xxxxx

{
  "mcp": {
    "dataflux-func-mcp-coding": {
      "enabled": true,
      "type"   : "remote",
      "url"    : "{DataFlux Func Access URL}/mcp/coding",
      "headers": {
        "Authorization": "Bearer atk-xxxxx:xxxxx"
      }
    },
  }
}

2. Practical Operation Example

First, confirm whether the AI Agent has successfully connected to DataFlux Func's MCP Coding service:

Then, proceed to input prompts normally.

After the AI Agent completes its task, refresh in DataFlux Func to see the code written by the AI.