Your AI agent doesn't know what your senior engineer knew.

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.

How it works

  • Analyze your PR history for tribal knowledge

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

  • MCP injects context before your agent writes a single line

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

  • Works in Cursor, Claude Code, and any MCP client

    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.

Real Examples

Your AI Agent doesn't know your tribal knowledge.

Real test: dotnet/runtime Issue #98259

Task: Remove where TKey : notnull from CountBy

Cursor
 Analyzes current code
 No PR history
 Doesn't know why constraints exist
 "Yes, you can delete it"
Cursor + KodHau
 Analyzes current code
 MCP gets your relevant tribal knowledge
         automatically and injects it in your agent
 Knows WHY decisions were made
 "Here's what breaks if you do"

Outcome

Cursor alone

"Yes, you can remove it from the signature"

Cursor + KodHau

  • PR #91507 — API contract was intentionally lazy
  • PR #106111 — CountByEnumerable refactor, don't touch return path
  • PR #122329 — Async overloads affected, update all variants

→ Don't just delete the line.

Give your AI agent the knowledge your best engineer never wrote down.

Join the waitlist and we’ll contact you. Book a demo to see KodHau MCP in action.