smixs/pohuy

What it solves

This project provides a set of plugins and configuration styles for AI agents (like Claude Code, Codex CLI, and Pi) to communicate using idiomatic, profane Russian. It replaces overly polite, verbose AI responses with concise, emotionally charged engineering slang, arguing that this style of communication is more direct and clearer for developers during high-stress technical incidents.

How it works

It functions as a "skill" or output style extension. It provides a mapping of technical situations to specific Russian swear words and idioms (a dictionary of 70+ idioms), a 10-step scale of project states (from "Triumph" to "Catastrophe"), and specific prompt instructions. These are integrated into the AI agent's system prompt or via hooks that reinforce the style throughout a session to prevent the AI from reverting to its default polite tone.

Who it’s for

Developers and engineers who prefer direct, unfiltered communication and find standard AI politeness inefficient or unrealistic in a technical context.

Highlights

• Multi-Agent Support: Compatible with Claude Code, OpenAI Codex CLI, and the Pi agent.
• Calibrated Intensity: Offers three levels of profanity (lite, full, ultra) to adjust the tone.
• Context-Aware Emotion: Matches the intensity of the swear words to the severity of the technical issue (e.g., using different terms for a minor bug vs. a total production crash).
• Engineering Mapping: Includes a specialized dictionary that maps technical concepts (like race conditions or flaky tests) to specific idiomatic expressions.
• Strict Rules: Maintains a boundary where code, commits, and security warnings remain professional and clean, while the profanity is reserved for the bugs and the system.