1,400 curated entries of cybersecurity vulnerabilities designed for training a Red Team GPT model. The dataset contains detailed records from 2021 to 2025, sourced from Exploit-DB, CVE details, and recent web sources like the CISA KEV catalog and The Hacker News. It was created by author hardik994 and last updated on the platform in June 2026.
Use Cases
- Training AI models for automated vulnerability research based on detailed vulnerability records.
- Supporting AI-driven penetration testing scenarios based on curated exploit entries.
- Analyzing vulnerability trends from 2021-2025 based on data aggregated from multiple sources.
Strengths
- Contains 1,400 curated entries providing a substantial foundation for model training.
- Aggregates data from multiple authoritative sources including Exploit-DB and the CISA KEV catalog.
- Covers a recent and defined time range from 2021 to 2025.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment for large-scale projects.
- Description metadata is limited; actual data quality requires manual inspection after download.
Provenance
- Source
- Exploit-DB, CVE details (nvd.nist.gov), CISA KEV catalog, The Hacker News
- Collection Method
- Curated aggregation from multiple web sources.
- Time Range
- 2021-2025
- Freshness
- Last updated 2026-06-20 11:12:10; freshness should be verified.