IFly Spark

Invokes the iFlytek SparkAgent task chain via MCP to upload files and drive SparkAgent workflows.

typescript

3

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "iflytek-ifly-spark-agent-mcp": {
      "command": "uv",
      "args": [
        "run",
        "ifly-spark-agent-mcp"
      ],
      "env": {
        "IFLY_SPARK_AGENT_APP_ID": "xxxx",
        "IFLY_SPARK_AGENT_BASE_URL": "xxxx",
        "IFLY_SPARK_AGENT_APP_SECRET": "xxxx"
      }
    }
  }
}

You can use the MCP server for the iFlytek SparkAgent platform to invoke its SparkAgent task chain from an MCP client. It exposes a tool to upload files and can be driven from local runtimes or remote MCP clients to automate SparkAgent workflows.

How to use

You interact with the MCP server through an MCP client. By default, a tool named upload_file is available, and it accepts a single required argument named file that should contain the path to the file you want to upload.

How to install

Prerequisites: you need an environment that provides the MCP runtime interface used by these examples. The server can be started using the MCP runtime tools uv or with the alternative transport sse.

Step 1. Start the MCP server locally with the default stdio transport.

uv run ifly-spark-agent-mcp

Step 2. If you prefer to use the SSE transport on a custom port, start the server like this.

uv run ifly-spark-agent-mcp --transport sse --port 8000

Step 3. You can also install and run the MCP server from the Git URL using uvx to fetch from GitHub. This pulls the server code and runs it directly.

uvx --from git+https://github.com/iflytek/ifly-spark-agent-mcp ifly-spark-agent-mcp

Additional sections

MCP client usage examples show how to enumerate available tools and invoke the upload_file tool. The client example uses Python and the stdio transport to connect to the local server, list tools, and call upload_file with a file path.

Environment variables shown for running the MCP server configure access to the iFlySpark Agent instance. You should provide these values in your environment when starting the server.

Available tools

upload_file

Uploads a file to the SparkAgent workflow by providing the file path through the file parameter.