Trendyol Cybersecurity: 53,202 Instruction-Tuning Examples for Defense

53,202 instruction-tuning examples for defensive cybersecurity, covering over 200 specialized domains including cloud-native threats and AI/ML security. Developed by the Trendyol Security Team and updated in July 2025, this dataset provides system, user, and assistant triplets for model training.

Use Cases

• Fine-tuning Large Language Models for incident response using the system/user/assistant instruction format
• Building defensive security assistants capable of addressing 200+ specialized domains
• Training models to recognize and mitigate cloud-native threats and AI/ML security vulnerabilities

Strengths

• 53,202 curated instruction-tuning examples
• Covers 200+ specialized cybersecurity domains
• Includes emerging topics like quantum computing risks and AI/ML security

Limitations

• Focus is restricted to defensive security, potentially excluding offensive security contexts
• Specific schema details for metadata columns are not provided in the summary

Provenance

Source

Trendyol Security Team

Collection Method

Curated

Freshness

Last updated July 2025.