KodHau MCP — The Governance Layer for your AI Agents

Overview

KodHau MCP gives your AI agent the tribal knowledge of your team: PR history, design decisions, and review comments your senior engineers never documented.

Before your agent writes a single line, KodHau MCP injects the tribal knowledge of your team: architecture, design decisions, constraints, rejected approaches, and review comments your senior engineers never documented.

KodHau scans closed PRs for architecture decisions, code review comments, rejected approaches, and diff context — the decisions your team made but never documented.

Key Features

• Your AI agent doesn't knowwhat your senior engineer knows.
• How it works
• Analyze your PR history for tribal knowledge
• MCP injects context before your agent writes a single line
• Works in Cursor, Claude Code, and any MCP client
• Real Examples
• Business
• Enterprise
• Your AI Agent doesn't know your tribal knowledge.
• Task: Remove where TKey : notnull from CountBy
• "Yes, you can remove it from the signature"
• Cursor + KodHau

Details

One tool: kodhau_get_wisdom. Your agent gets constraints, architecture decisions, and tribal knowledge for that exact file and function — before it touches anything.

KodHau is an MCP server that runs locally. Your code and PR data never leave your machine. Compatible with any client that supports the Model Context Protocol.