kagi-summarizer
Warn
Audited by Socket on Mar 27, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: The skill's purpose and data flow are mostly coherent and it uses Kagi's official API, but the installation model introduces notable trust risk. It downloads and executes a mutable prebuilt binary from a personal GitHub publisher rather than Kagi, and that code receives the user's KAGI_API_KEY. Public source and release history reduce the likelihood of outright malware, but the third-party supply-chain and credential-forwarding footprint is broader than ideal for a simple summarizer wrapper.
Confidence: 86%Severity: 66%
Audit Metadata