🤖 AI Prompts Cheatsheet
Daniel Osei — AI-Assisted Security Engineer
Why AI Changes the Game for Threat Intelligence Analysis
In the age of rapidly evolving cyber threats, utilizing AI tools like ChatGPT and Claude can significantly enhance threat intelligence workflows. These tools can help analysts sift through vast amounts of data, summarize findings, and generate threat reports efficiently, allowing teams to respond faster and more effectively.
Before You Start: How to Set Context Properly
Setting the right context is crucial for maximizing AI output. Context acts as the foundation for AI responses, influences the relevance of generated content, and determines prompt effectiveness. To ensure your AI tool understands your requirements:
- Be Specific: Detail exactly what you need. Include relevant parameters, timeframes, and types of threats.
- Define Roles: Assign roles in your prompts (e.g., “You are a cybersecurity analyst…”) to tailor the responses.
- Provide Examples: If applicable, show examples of desired outputs to guide the AI effectively.
Core Prompts Cheatsheet
What it does: Summarizes recent threats.
When to use it: Weekly reports or briefing preparation.
How to customize: Specify particular threat types or sources if necessary.
What it does: Analyzes TTPs linked to specific attacks.
When to use it: When gathering intelligence on current malware and exploits.
How to customize: Mention specific ransomware families for better context.
What it does: Provides a structured outline for reporting.
When to use it: Before drafting threat reports.
How to customize: Specify the threat actor in question.
What it does: Provides a ranked list of critical vulnerabilities.
When to use it: For vulnerability management prioritization.
How to customize: Adjust the number of vulnerabilities listed as needed.
What it does: Provides insights into the impacts of security incidents.
When to use it: For incidents response review or awareness sessions.
How to customize: Specify sectors or regions for relevance.
What it does: Lists critical IOCs for detection and prevention.
When to use it: For fortifying detection methods.
How to customize: Mention specific malware types for better focus.
Weak vs Strong Prompt Examples
Advanced Prompt Techniques
To further enhance your interactions with AI tools, consider these advanced prompting techniques:
- Role Prompting: Assign AI specific roles to contextualize requests.
- Chain-of-Thought: Encourage the AI to break down complex queries into manageable parts.
- Few-Shot Examples: Provide sample responses to train AI on your expected format.
- Output Formatting: Specify how you want the output (e.g., bullet points, tables).
Claude vs ChatGPT: Which Works Better For This
Both Claude and ChatGPT offer robust capabilities, but with slight differences. ChatGPT may provide more concise responses, whereas Claude can produce more detailed explanations. For tabulated data, Claude’s outputs are typically more structured. Testing both tools on your specific prompts could yield insights into which produces better results for your unique workflows.
Tips for Getting Consistent Results
For optimal AI interactions, adhere to these strategies:
- Set Context: Always provide rich context to avoid vague responses.
- Be Specific: Use clear and concise requests to limit misinterpretation.
- Iterate: Don’t hesitate to refine prompts based on received outputs.
- Feedback Loop: Provide feedback on outputs to guide future AI performance.
Quick Reference: All Prompts in One Place
- Generate a summary of emerging cyber threats reported in the last week.
- What TTPs are associated with the recent ransomware attacks?
- Create a detailed threat report outline for an APT group.
- List the top 10 vulnerabilities currently exploited in the wild.
- Summarize notable security breaches from the last month.
- Identify key indicators of compromise for the latest malware family.