감사 이력
senior-data-scientist - 8 감사
버전 비교
감사된 버전 간의 기능 및 발견 항목 변화, 최신순.
| 버전 | 날짜 | 결과 | 검토 항목 | 이전 버전 대비 변화 |
|---|---|---|---|---|
| v8 최신 | 9 jul 2026, 11:16 | 확인된 발견 항목 없음 | 1 | 기능 변화 없음 |
| v7 | 9 jul 2026, 11:16 | 확인된 발견 항목 없음 | 1 | 기능 변화 없음 |
| v6 | 6 jul 2026, 02:42 | 확인된 발견 항목 없음 | 1 | 기능 변화 없음 |
| v5 | 6 jul 2026, 02:42 | 확인된 발견 항목 없음 | 1 | 기능 변화 없음 |
| v4 | 28 jun 2026, 08:29 | 1 확인됨 | 0 | 기능 변화 없음 |
| v3 | 16 ene 2026, 16:55 | 확인된 발견 항목 없음 | 0 | 기능 변화 없음 |
| v2 | 16 ene 2026, 16:55 | 확인된 발견 항목 없음 | 0 | 외부 명령어 스크립트 포함 |
| v1 | 15 ene 2026, 11:41 | 확인된 발견 항목 없음 | 0 | 기준선 |
9 jul 2026, 11:16
No evidence of malware, prompt injection, network reconnaissance, or data exfiltration was found in the reviewed files. Most static findings are false positives caused by Markdown code fences or inline references. One documented command block includes state-changing deployment examples that should require explicit user review.
기능 검토 항목 (1)
이 스킬에 대해 예상될 수 있는 실제 로컬 기능이므로 검토가 필요하지만, 확인된 악의적 동작으로는 집계되지 않습니다.
위험 요인
9 jul 2026, 11:16
No evidence of malware, prompt injection, network reconnaissance, or data exfiltration was found in the reviewed files. Most static findings are false positives caused by Markdown code fences or inline references. One documented command block includes state-changing deployment examples that should require explicit user review.
기능 검토 항목 (1)
이 스킬에 대해 예상될 수 있는 실제 로컬 기능이므로 검토가 필요하지만, 확인된 악의적 동작으로는 집계되지 않습니다.
위험 요인
6 jul 2026, 02:42
No evidence of malware, prompt injection, network reconnaissance, or data exfiltration was found in the reviewed files. Most static findings are false positives caused by Markdown code fences or inline references. One documented command block includes state-changing deployment examples that should require explicit user review.
기능 검토 항목 (1)
이 스킬에 대해 예상될 수 있는 실제 로컬 기능이므로 검토가 필요하지만, 확인된 악의적 동작으로는 집계되지 않습니다.
위험 요인
6 jul 2026, 02:42
No evidence of malware, prompt injection, network reconnaissance, or data exfiltration was found in the reviewed files. Most static findings are false positives caused by Markdown code fences or inline references. One documented command block includes state-changing deployment examples that should require explicit user review.
기능 검토 항목 (1)
이 스킬에 대해 예상될 수 있는 실제 로컬 기능이므로 검토가 필요하지만, 확인된 악의적 동작으로는 집계되지 않습니다.
위험 요인
28 jun 2026, 08:29
AI review found the static weak-cryptography and network-reconnaissance findings to be false positives caused by data science and systems terminology. The skill includes documented shell commands and placeholder Python scripts, but no evidence of malicious execution, exfiltration, network access, or prompt injection was found.
확인된 보안 우려 사항 (1)
정적 오탐 무시됨 (2)
이러한 정적 일치는 의미론적 검토에서 기각되었거나 스키마 전용 토큰과 일치했으므로 투명성을 위해 표시되지만 품질 점수에는 영향을 주지 않습니다.
위험 요인
⚙️ 외부 명령어 (2)
16 ene 2026, 16:55
All 74 static findings are false positives. Scanner misinterpreted markdown headings (# symbol) as weak cryptographic algorithms, bash code examples as command injection, and documentation references as C2/network reconnaissance. The skill contains legitimate data science documentation and Python boilerplate for experiment design, feature engineering, and model evaluation. No actual cryptographic implementations, subprocess calls, or network operations exist.
위험 요인
16 ene 2026, 16:55
All 74 static findings are false positives. Scanner misinterpreted markdown headings (# symbol) as weak cryptographic algorithms, bash code examples as command injection, and documentation references as C2/network reconnaissance. The skill contains legitimate data science documentation and Python boilerplate for experiment design, feature engineering, and model evaluation. No actual cryptographic implementations, subprocess calls, or network operations exist.
위험 요인
15 ene 2026, 11:41
All 48 static findings are false positives. Scanner misinterpreted documentation about security practices as cryptographic code and markdown code block syntax as command execution. No actual cryptographic implementations, subprocess calls, or network operations exist in the codebase.