smixs/pohuy

Режим идиоматического русского мата для AI-агентов. Короче, душевнее, эффективнее. 18+

What it solves

This project provides a set of plugins and style guides for AI agents (like Claude Code, Codex CLI, and Pi) to communicate in a highly expressive, idiomatic Russian slang/profanity-laden style. It aims to replace overly polite, verbose AI responses with concise, emotionally charged, and technically accurate summaries of software engineering problems, arguing that such a style is more efficient for developers.

How it works

It functions as a "skill" or output style extension. It provides a dictionary of over 70 engineering-mapped idioms, a 10-step scale of project states (from "Triumph" to "Catastrophe"), and specific prompt instructions. These are integrated into AI agents via plugins, shell scripts that modify settings files, or native extensions, ensuring the AI maintains this specific persona across sessions.

Who it’s for

Software engineers and developers who prefer blunt, idiomatic Russian communication over standard AI politeness, and who find that high-emotion, low-jargon summaries of bugs and crashes are more immediately clear.

Highlights

  • Multi-Agent Support: Compatible with Claude Code, OpenAI Codex CLI, and the Pi agent.
  • Calibrated Intensity: Offers three levels of intensity (lite, full, ultra) to control the amount of profanity.
  • Engineering Mapping: Maps specific technical failures (e.g., race conditions, flaky tests) to corresponding idiomatic expressions.
  • Emotional Calibration: Adjusts the tone based on the severity of the project state, ensuring the profanity is directed at the bugs, not the user.
  • Detailed Ontology: Includes a specialized lexicon for agentic behavior and software failures.

Related

  • Project
    tanweai/pua

    pua is a multi‑platform skill that injects corporate‑style “performance‑improvement‑plan” rhetoric and a 7‑point debugging checklist into AI coding assistants (Claude Code, OpenAI Codex, Cursor, Kiro, etc.). It auto‑triggers when the model shows lazy patterns (retries, blaming the user, passive waiting) or can be invoked manually with `/pua`. The skill forces the model to exhaust solutions, verify results, and act proactively, offering language‑specific packs (Chinese, English, Japanese). Benchmarks on a handful of bugs show higher fix and verification counts, though speed gains are modest. Installation is via the respective tool’s plugin/skill system (e.g., `npx skills add tanweai/pua --skill pua-en` for Claude Code).

  • Project
    yzhao062/agent-style

    A set of 21 professional writing rules and integration tools that prevent AI agents from using common LLM prose patterns in technical documentation.

  • Dispatch
    i-have-adhd: Optimizing Coding Agent Outputs for Executive Dysfunction

    i-have-adhd is a set of rules and plugins for coding agents that eliminates verbosity and preamble, forcing AI assistants to lead with actions and concrete next steps.

  • Project
    alchaincyf/huashu-skills

    huashu‑skills is a curated index of 57 Claude‑compatible *Agent Skills* (design, data, video, persona, etc.). 22 lightweight skills live in this repo; the rest are separate repos. Install with a one‑line `git clone` (or copy a sub‑folder), restart the agent, and the skill becomes usable. An included updater checks for newer versions. The collection provides ready‑made, plug‑in functionality for AI agents without writing code.

  • Project
    Cjbuilds/Codex-Orchestration

    Codex Orchestration is a Codex plug‑in that lets you attach other LLMs (Claude Fable 5, Opus 5, OpenRouter models, etc.) to a task and assign them specific roles—Planner, Advisor, Designer, Executor. Codex stays the top‑level orchestrator, iterating plans, reviewing them, optionally designing UX artefacts, and finally executing code. The plug‑in is installed via the Codex marketplace, configured with skill prompts inside a Codex chat, and supports role‑specific effort levels. It aims to improve plan quality, enable parallel work, and reduce premium‑model usage. MIT‑licensed.