Coding Standards

This is a linter-as-context for AI code generation. It injects coding standards into your LLM's prompts before it writes code, enforcing patterns like DDD layers, SOLID principles, dependency injection, custom exceptions, and max line counts per method. Works via stdio with Cursor, VS Code, Claude Desktop, and other MCP clients. The value prop is reducing the AI slop that technically runs but fails code review. Instead of fixing generated code after the fact, you get repository interfaces, proper error types, and test scaffolding from the first output. Configure once in your MCP settings, optionally override rules in `.corbat.json`, and it auto-detects your stack to apply relevant guardrails.