Mark Levison avatar

Mark Levison

Community

@mlevison · Ottawa, On

5Followers
|
23Public Repos
|
3Published Skills

Mark Levison is a Certified Scrum Trainer who has been helping teams Agile, Scrum and Kanban style approaches since 2001.

Skills Distribution
DomainBusiness, Fi...Critical Reasoning.. (35%)Systems Thinking &.. (30%)Model Output Verif.. (25%)Agile & Scrum Coac.. (10%)

Agent Skills by Mark Levison

Showing 3 vetted skills indexed across 1 GitHub repositories.

Frequently Asked Questions About Mark Levison

FAQPage Schema
What tasks can I accomplish with Mark Levison's skills?▼

You can stress-test claims and arguments one question at a time, verify model-generated text with eleven interrogation probes to catch hallucinations, and map recurring problems to their root causes and feedback loops through guided questioning rather than prescribed solutions.

Who are these skills designed for?▼

They target knowledge workers, agile teams, and decision-makers who need to validate reasoning before acting. The author is a Certified Scrum Trainer practicing Agile, Scrum, and Kanban coaching since 2001, publishing under agilepainrelief.com.

How do I use the critical-thinking-genai skill in practice?▼

Invoke it when you paste model output you are about to rely on, suspect a hallucination, or discover an earlier answer in the conversation was wrong. It applies eleven probes that test the output itself, since there is no author to walk back through the reasoning.

What license and cost apply to these skills?▼

All three new skills are released under the Creative Commons CC BY-SA 4.0 license, making them free to use and share with attribution and share-alike conditions. They are authored by agilepainrelief.com at versions 0.1 and 0.35.

When should I use critical-thinking versus critical-thinking-genai?▼

Use critical-thinking for human-authored claims, decisions, or arguments you want poked holes in. If the text came from a model, the skill explicitly hands off to critical-thinking-genai, which is purpose-built for output with no reasoning author.