Self-hosted MCP bridge between Claude Chat and Claude Code

kOlapsis/herald: Self-Hosted MCP Bridge for Claude AI

For developers leveraging Anthropic's Claude AI, managing interactions between Claude Chat and Claude Code can be complex, especially in self-hosted environments. The kOlapsis/herald project addresses this by providing a direct, open-source MCP (Message Communication Protocol) bridge. This tool enables seamless communication between Claude Chat and Claude Code instances that you host yourself, streamlining your AI development workflow. It's designed for technical users who need precise control over their AI agent interactions and data.

What it Does

kOlapsis/herald acts as an intermediary, allowing your self-hosted Claude Chat agent to send prompts and receive responses from your self-hosted Claude Code agent. This facilitates scenarios where a conversational AI can generate code, and a code-focused AI can execute or analyze that code, all within your controlled infrastructure. The bridge translates messages between the two Claude instances, ensuring compatibility and efficient data flow.

Key Features

• Self-Hosted Operation: Run the entire bridge and your Claude instances locally or on your private servers, ensuring data privacy and security.
• MCP Protocol Support: Leverages the Message Communication Protocol for robust and standardized agent communication.
• Claude Chat Integration: Connects to your Claude Chat instance for conversational input and output.
• Claude Code Integration: Connects to your Claude Code instance for code generation, analysis, and execution tasks.
• Open-Source: The project is available on GitHub, allowing for inspection, modification, and community contributions.

Who it's For

This tool is specifically for AI developers and engineers who are:

• Building custom AI agents and workflows using Anthropic's Claude models.
• Requiring a self-hosted solution for enhanced data control and privacy.
• Integrating conversational AI capabilities with code generation or analysis tools.
• Familiar with setting up and managing AI infrastructure and communication protocols.
• Looking for an open-source solution to customize and extend their AI agent ecosystem.