NYC
skills/smithery/ai/gog/Gen Agent Trust Hub

gog

Fail

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: HIGHEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS] (HIGH): The skill installs an external binary gogcli from a non-trusted Homebrew tap (steipete/tap). This source is not on the trusted organization list.
  • [DATA_EXFILTRATION] (MEDIUM): The skill accesses highly sensitive user data including Gmail messages, Drive files, and Calendar events. It includes capabilities to send emails and export documents, which are potential exfiltration vectors.
  • [PROMPT_INJECTION] (LOW): The skill is vulnerable to Indirect Prompt Injection (Category 8). It ingests untrusted data from Gmail and Docs which could contain malicious instructions designed to trigger its write capabilities. Evidence Chain: 1. Ingestion points: gog gmail search, gog docs cat, gog drive search. 2. Boundary markers: Absent. 3. Capability inventory: gog gmail send, gog sheets update, gog calendar create. 4. Sanitization: Absent.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 17, 2026, 06:37 PM