Verified project record

KKKKhazix/khazix-skills

khazix-skills is a collection of practical, tested structured instruction sets (Agent Skills) that can be directly loaded by 40+ AI agents to automate tasks such as goal setting, disk cleaning, documentation alignment, research, and writing. Users provide vague ideas or triggers and the skills guide the agent through producing structured outputs including goal documents, interactive HTML disk reports, project documentation updates, and PDF research reports.

Project overview

It transforms multi-step agent workflows—from goal definition and disk cleanup to deep research and style-matched writing—into ready-to-use skills loadable by 40+ AI agents without requiring coding, lowering the barrier for automating complex tasks.

Project type: AI Agent · AI Coding
Use cases: Coding & Development
Deployment: Refer to project documentation
License: MIT

Best for

Developers and general users who want pre-built, tested agent skills for tasks like goal definition, disk cleanup, documentation alignment, deep research, and style-matched writing without writing code.

Key capabilities

Transforms a vague idea into a clear goal task document that can be fed into an agent's goal mode for autonomous execution.
Scans the entire disk and generates an interactive HTML report with a three-color grading system for cleanup decisions.
Allows agents to fetch daily AI news and updates using natural language without needing an API key.
Aligns session changes with project documentation, agent memory, and rules, providing a change summary.
Conducts deep product/company research by combining historical and horizontal competitive analysis into a PDF report.
Enables the agent to write long-form articles in a specific author's tone and rhythm.

Limitations and risks

Windows code for the storage-analyzer skill is ready but not fully tested.
When using the Leader skill, agents may find unintended shortcuts if goals lack constraints (Harness). Users should provide adequate constraints alongside goals.

Getting started

Setup is rated easy: tell a compatible agent to install a skill by providing its GitHub path, for example: https://github.com/KKKKhazix/khazix-skills/tree/main/<skill-name>. Alternatively, manually paste the skill content into the agent. No coding is required.

Evidence and sources

Open original GitHub project