skills/mukul975/anthropic-cybersecurity-skills/building-attack-pattern-library-from-cti-reports/Gen Agent Trust Hub
building-attack-pattern-library-from-cti-reports
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were detected. The skill performs legitimate threat intelligence analysis tasks using well-documented libraries.
- [EXTERNAL_DOWNLOADS]: The skill references MITRE's official ATT&CK STIX dataset hosted on GitHub. This is a well-known and trusted source for cybersecurity data, essential for the skill's purpose.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted CTI reports. 1. Ingestion points: scripts/agent.py reads report files via CLI. 2. Boundary markers: None present. 3. Capability inventory: re, json, stix2, datetime, collections (no high-risk capabilities like exec, eval, or shell access). 4. Sanitization: Extraction is limited to structural regex matching. The risk is evaluated as safe.
Audit Metadata