Nice AI: Cybersecurity Dataset with CVE Records, Exploits, and Threat Intelligence

A collection of cybersecurity data from 1999 to 2025, including approximately 300,000 CVE records. The dataset aggregates disclosed reports from HackerOne, exploits from ExploitDB, red team prompts, and structured threat intelligence from MITRE ATT&CK. It was created by Zain Ali and last updated on HuggingFace in May 2026.

Use Cases

• Training vulnerability classification models based on CVE records and software vulnerability JSONL.
• Generating or analyzing red team prompts based on the included red team prompts dataset.
• Mapping attack techniques and procedures using the integrated MITRE ATT&CK framework data.
• Developing exploit detection systems based on ExploitDB exploits and shellcodes.
• Analyzing bug bounty program scope and trends using the HackerOne and Bugcrowd scope data.

Strengths

• Includes approximately 300,000 CVE records spanning from 1999 to 2025.
• Aggregates data from multiple established sources like HackerOne, ExploitDB, and MITRE ATT&CK.
• Covers diverse cybersecurity concepts including exploits, templates, and structured threat intelligence.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count and file size are unknown, which may limit suitability assessment.
• The description references a full page for details, suggesting metadata here is incomplete.

Provenance

Source

HuggingFace user bountyhunterzain (Zain Ali).

Collection Method

Aggregated from multiple open cybersecurity sources.

Time Range

1999-2025

Freshness

Last updated 2026-05-06 10:17:04; freshness should be verified.

Geography

null