open-code-review
Warn
Audited by Socket on Oct 28, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The core capability is coherent: a code-review skill invoking Alibaba's official OCR CLI with expected code/LLM data flows. Main risk comes from supply-chain and trust-extension behavior: curl-to-shell from raw GitHub, unpinned command-file downloads, and especially transitive installation of another skill from a third-party GitHub repo unrelated to Alibaba. No strong evidence of credential theft, covert exfiltration, or confirmed malware.
Confidence: 90%Severity: 64%
Audit Metadata