KKKKhazix/khazix-skills

What it solves

Khazix Skills is a collection of structured instruction sets (Skills) designed to extend the capabilities of AI agents. It solves the problem of repetitive prompting and the lack of specialized workflows for common tasks like goal definition, system cleanup, research, and content creation, allowing users to integrate professional-grade behaviors into their AI agents without writing complex prompts from scratch.

How it works

The project follows the Agent Skills open standard, providing SKILL.md files that can be directly loaded by compatible agents (such as Claude Code, Codex, Cursor, and Kimi Code). These files act as structured guidelines that the AI agent follows to execute specific tasks. If an agent does not support the standard, the instructions can be manually pasted into the conversation as project rules.

Who it’s for

It is intended for users of AI agents who want to automate specialized tasks, improve the consistency of their AI's output, or implement specific professional workflows (like deep research or a specific writing style) within their development or productivity environment.

Highlights

• leader: Transforms vague ideas into a detailed, actionable goal document for long-running AI tasks, focusing on constraints and verification.
• storage-analyzer: Scans local disks (Mac/Windows) and generates an interactive HTML report with a three-color priority system for safe file cleanup.
• aihot: Integrates real-time AI news and trends from aihot.news without requiring an API key.
• neat-freak: Synchronizes project documentation, agent memory, and rules (like CLAUDE.md) after a task is completed to prevent "knowledge rot."
• hv-analysis: Performs deep research on companies or concepts using a longitudinal (historical) and lateral (competitive) analysis method to produce long-form PDF reports.
• khazix-writer: Implements a specific, opinionated writing style for long-form articles, avoiding corporate jargon and generic AI phrasing.