khazix-skills: a collection of standardized instruction sets to extend AI agent capabilities for system utility, research, and writing

What it solves

Khazix Skills is a collection of structured instruction sets (Skills) designed to extend the capabilities of AI agents. It provides a set of specialized tools for common tasks such as system cleanup, AI news aggregation, project documentation maintenance, deep research, and stylized writing, which can be directly loaded by agents that follow the Agent Skills open standard.

How it works

The project provides SKILL.md files that act as structured instructions for AI agents. These can be installed via a simple command in compatible agents (like Claude Code, Codex, Cursor, etc.) or manually by providing the SKILL.md file as a project rule or prompt.

Each skill implements a specific workflow:

• storage-analyzer: Scans disk space and generates an interactive HTML report with a three-color priority system (Green/Yellow/Red) for cleanup decisions.
• aihot: Fetches daily AI news and reports from aihot.virxact.com without requiring an API key.
• neat-freak: Synchronizes project documentation, agent memory, and rule files (like CLAUDE.md) after a task is completed to prevent "knowledge rot."
• hv-analysis: Performs a vertical (historical evolution) and horizontal (competitor comparison) analysis to generate comprehensive PDF research reports.
• khazix-writer: Applies a specific writing style, rhythm, and set of banned words to generate long-form content.

Who it’s for

Users of AI agents (such as Claude Code, Codex, Cursor, and others) who want to add specialized, high-utility functions to their AI assistants without writing complex prompts from scratch.

Highlights

• Standardized Format: Follows the Agent Skills open standard, making it compatible with 40+ different agents.
• Safe Disk Cleanup: The storage analyzer uses a read-only scan and requires browser-based confirmation for any deletions.
• AI News Integration: Direct access to curated AI news streams and reports.
• Project Hygiene: The neat-freak skill ensures that documentation and agent memory stay aligned with the actual code changes.
• Deep Research: Capable of generating 10,000 to 30,000 word PDF reports using a specific analysis methodology.