Works with the AI Agents you already use
crg sensitivity analyst
Socratic coach for students learning the CyberRiskGuardian cybersecurity risk assessment method in English and French.
Free
crg sensitivity analyst
Example session with this skill installed
I'm stuck on my risk scenario for the hospital lab case. I put 'data breach' as the scenario but my instructor says it's just a threat. What's the difference?
- Read your context and instructions
- Compiled the crg sensitivity analyst
I can help you bridge that gap. Identifying the threat is a good starting point, but we need to build the full chain.
Where you are
You have correctly identified the "what"—a data breach is indeed the core threat actor's objective in this case.
Concept in plain words
Think of a "threat" as a specific hazard, like a "storm." A "scenario" is the story of how that storm hits your specific house. It includes how the wind got in (the weakness), which room it wrecked (the asset), and how much it costs to fix (the consequence). A scenario needs a beginning, a middle, and an end.
Analogy limit: Unlike a real storm, cyber scenarios usually involve a human intent or a specific system failure you can actually map out.
A question for you
Looking at the hospital lab case, what is the specific "crown jewel" data asset that, if breached, would stop the doctors from treating patients immediately?
Check on your reasoning
| Your statement | What holds | Question to test it |
|---|---|---|
| "Data breach" | It identifies the threat category. | Which specific vulnerability in the lab's portal allows this? |
| [Missing link] | The asset is undefined. | Is it patient names, or the actual test results? |
Watch out for
A threat name is not a scenario. Avoid just naming the "what" (breach, ransomware, fire). You must describe the path from the threat source to the business impact.
Your next step
Draft a single sentence using this path
**Threat Source** (who) ->
**Weakness** (how) ->
**Asset** (what) ->
Consequence (impact). Paste it here for feedback.
Next steps
- Identify one specific vulnerability mentioned in the case brief.
- Name the primary business process that fails if that data is gone.
- Ask me to quiz you on the "Six Parameters" once your scenario is ready.
Connects securely to your tools. The creator never sees your data.
About this skill
Students often treat cybersecurity risk assessments like a math problem rather than a logic puzzle. They struggle to differentiate between threats and scenarios, or they fail to align risk appetite with business consequences. This skill provides a Socratic coaching layer that guides students through the CyberRiskGuardian (CRG) method without doing the work for them. It forces students to think through the causal chains of risk while preventing the academic integrity issues common with generative AI.
What it does
- Socratic coaching asks guiding questions to help students discover logic flaws in their own risk assessments.
- Concept clarification explains complex parameters like residual risk vs. tolerated risk using plain language and analogies.
- Reasoning validation provides a structured table checking the student's logic against common mistakes in the CRG method.
- Practice quizzes generates interactive test sessions on risk indices, CVSS limits, and KRIs.
- Multilingual support offers full coaching and technical terminology in both English and Québec-specific professional French.
How it works
- Context check identifies the student's request type and verifies if the instructor permits AI assistance for the task.
- Logic analysis compares student drafts or questions against established CRG concepts and common pedagogical pitfalls.
- Feedback delivery outputs a structured response starting with what the student got right before pointing out the next logical step.
- Iterative guidance provides bigger hints or fictional examples only after the student remains stuck through multiple attempts.
Frameworks & tools
Built specifically for the CyberRiskGuardian (CRG) risk method. It uses structured references for the CRG ontology, markers for analytical estimates, and adheres to academic integrity protocols for university-level cybersecurity courses.
Why this beats prompting it yourself
Generic AI models often default to writing the assignment for the student, which leads to academic misconduct and zero learning. This skill is hard-coded to refuse deliverable generation, forcing students to apply the method themselves while using university-vetted analogies and terminology.
Use cases
- Draft feedback checking a student's scenario description for missing causal links before they submit.
- Exam prep running a 5-question quiz on the difference between risk appetite and tolerance.
- Concept hurdles explaining the specific limitations of CVSS scores in a business risk context.
- Stuck resolution helping a student identify which "crown jewel" asset is affected when they can only name a threat.
Known limitations
This skill will not calculate final risk indices, produce model answers, or write graded memos. It does not provide legal advice or professional audit opinions for real-world organizations.
Will not generate graded deliverables, final risk numbers, or CVSS scores. Restricted to CRG method terminology. Requires manual confirmation of instructor AI policies.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before any payment.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 7 days ago
- Free with an Agensi account