kudosflow

kudosflow is a Flowise-like LangChain extension that brings an AI flow management experience directly into your VSCode editor. It uses SceneGraphManager to visualize, design, and execute LangChain-based AI flows, and exports the flows as JSON that SceneGraphManager can execute. You can import Flowise-compatible JSONs, edit or test flows inside VSCode using the built-in ReactFlow-based editor, and then run them to orchestrate LLM components, memory, embeddings, and vector stores in a directed acyclic graph. When interacting with the integrated assistant, you can highlight code and chat with the AI to understand, refactor, or comment code, with conversations saved to history for export.

To use kudosflow, install the extension in VSCode, then leverage the JSON-based AI flows to design your pipelines. Designers can switch between different flows by loading different JSON definitions, enabling rapid deployment of various LangChain configurations without writing code. The extension coordinates with SceneGraphManager under the hood to execute the defined nodes in the correct order, making complex AI workflows accessible even to developers who aren’t AI specialists.

Prerequisites:

• Visual Studio Code installed
• Node.js (recommended LTS) and npm
• Internet access to install dependencies

Step-by-step installation:

1. Clone the repository or open the project in VSCode git clone https://github.com/akudo7/kudosflow.git cd kudosflow

2. Install dependencies npm install

3. Build or prepare the extension (if applicable in your setup) npm run build

4. Install the extension in VSCode

   • Open VSCode, go to Extensions, and either search for kudosflow in the marketplace or load the extension from the repository using VSCode’s extension management.
   • If you’ve built a VSIX locally, you can install it via: code --install-extension kudosflow-*.vsix

5. Run the MCP server entry (if you’re running a local server component) npm run start

Prerequisites recap: ensure Node.js and npm are installed, VSCode is up to date, and you have network access to fetch dependencies and marketplace extensions.

Tips and notes:

• The kudosflow token for pre-release usage is valid until March 31, 2026. Store tokens securely in VSCode Settings under kudosflow > token.
• kudosflow can import and execute Flowise-compatible JSON flows; use Flowise JSONs to design AI flows visually and then export/import them into kudosflow.
• Environment variables and extension settings (e.g., token, paths for flows, and editor preferences) can customize the editor experience and integration points with SceneGraphManager.
• If you encounter issues, verify that the Flowise-compatible JSONs are valid and that the SceneGraphManager is properly initialized in your environment. Check the Settings panels (token, temp folders, and Copilot/LangChain settings) for proper values.
• The extension provides a chat interface in VSCode that uses the configured LLMs (OpenAI, Anthropic, Azure OpenAI, Ollama) and supports memory, embeddings, and vector stores through LangChain nodes.
• When importing or switching JSON flows, ensure dependencies (like Weaviate or Qdrant for vector stores) are available in your environment if your flows rely on them.