Integrate Model Context Protocol (MCP) into Flow API - 100,000 GURU

[evahteev]

Overview

The Model Context Protocol (MCP) enables applications to provide standardized context for Language Model (LLM) interactions, clearly separating the responsibilities of context management and LLM processing. The Python SDK implementing MCP simplifies the creation of:

• MCP clients capable of connecting to any MCP-compliant server.
• MCP servers that expose data resources, prompts, and tools.
• Standardized transports such as stdio and Server-Sent Events (SSE).
• Complete handling of MCP protocol messages and lifecycle events.

Objective

This issue aims to integrate MCP support into the Flow API of the Guru Framework. The goal is straightforward: any application built using the Guru Framework should seamlessly expose its data resources and business workflows to agentic interfaces (e.g., Guru Framework, Claude Desktop), including blockchain data managed through the Flow Data Management platform and BPMN-defined workflows.

Acceptance Criteria

• Extend Flow API to serve as an MCP-compliant server, feel free to use https://github.com/modelcontextprotocol/python-sdk.
• Support standardized data exposure through MCP.
• Enable seamless interaction between Guru Framework applications and LLM agentic interfaces.
• Demonstrate functional interaction using at least one standard transport (stdio or SSE).
• Complete documentation of MCP integration, including setup and usage examples.

Reward

This task is part of our experimental reward-driven initiative. Upon successful PR merge and team review, contributors will receive:

🎉 100,000 GURU tokens

Getting Started

• Familiarize yourself with Model Context Protocol specification.
• Explore the current Flow API and understand its capabilities.

Additional: https://www.youtube.com/watch?v=lbyPJqCI-tw

[evahteev]

Feel free to reach for additional info in Guru Chat: https://t.me/dexguru
and Discord: https://discord.gg/dPW8fzwzz9