Tuba Workflow

Exposes Tuba.ai workflow control as MCP tools for running, monitoring, and updating your vision pipelines.

python

GitHub Stars

python

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": {
    "devisionx-tuba-workflow-mcp-server": {
      "command": "uv",
      "args": [
        "run",
        "--directory",
        "<absolute_path_to_tuba_workflow_mcp_server_folder>",
        "python",
        "tuba_workflow_mcp_server.py"
      ],
      "env": {
        "TUBA_WORKFLOW_ACCESS_TOKEN": "<your_access_token>"
      }
    }
  }
}

You run this MCP server to expose your Tuba.ai workflow as programmable tools. It lets other apps, scripts, or assistants control and monitor your AI vision pipelines through a standard MCP interface, making it easy to automate tasks like running workflows, checking status, and retrieving results.

How to use

You interact with the server through an MCP client or a workflow-enabled assistant. Start a local MCP session, then issue high-level requests such as starting your Tuba workflow, checking progress, fetching results, inspecting the current workflow block configuration, or updating parameters and uploading files to blocks. The tools provide concrete actions you can invoke to manage your workflow end-to-end without manual GUI interactions.

How to install

Prerequisites include Python 3.10 or higher and the uv package manager for Python environments.

Clone the project directory and enter it.

git clone <this_repo>
cd <this_repo>

Install uv, a fast Python package installer and environment manager.

curl -LsSf https://astral.sh/uv/install.sh | sh

Create a virtual environment and install dependencies.

uv venv
uv pip install -r requirements-lock.txt

Local development and testing

Configure a local access token for your Tuba workflow and load it from an environment file during development.

# Create an environment file at project root
touch .env

TUBA_WORKFLOW_ACCESS_TOKEN="your_secret_token_goes_here"

Claude Desktop configuration

If you use Claude Desktop, set up the MCP server entry in your Claude configuration so you can invoke the server from chat messages.

{
  "mcpServers": {
    "TubaWorkflow": {
      "command": "<home_path>/.local/bin/uv",
      "args": [
        "run",
        "--directory",
        "<absolute_path_to_tuba_workflow_mcp_server_folder>",
        "python",
        "tuba_workflow_mcp_server.py"
      ],
      "env": {
        "TUBA_WORKFLOW_ACCESS_TOKEN": "<your_access_token>"
      }
    }
  }
}

Start the MCP server locally

Run the server in a development session to test and iterate.

uv run python tuba_workflow_mcp_server.py

Verification

After starting, verify you can connect with an MCP client or Claude Desktop and issue basic operations like run, status, and result.

Available tools

run

Executes the workflow for your project, starting the Tuba pipeline and returning control to the MCP client.

status

Retrieves the current status of the running workflow, including progress and any active steps.

result

Fetches the results of the workflow execution. If outputs are files, they are provided as result.zip.

get_workflow_blocks

Fetches the current configuration of all workflow blocks so you can review block IDs and parameters.

update_workflow_blocks_data

Updates parameters and uploads files to workflow blocks. Supports local files, remote URLs, or base64-encoded data.

Built by

VeilStrat

AI signals for GTM teams