skills/openclaw/skills/gog/Gen Agent Trust Hub

gog

Warn

Audited by Gen Agent Trust Hub on Mar 18, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the gog CLI utility via a third-party Homebrew tap (steipete/tap/gogcli).\n- [DATA_EXFILTRATION]: The skill enables comprehensive access to sensitive user information across Gmail, Drive, Contacts, and Sheets. It also references authentication processes involving local client_secret.json credential files.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from external emails and documents.\n
  • Ingestion points: Untrusted content is retrieved and placed into the agent's context through commands like gog gmail search, gog docs cat, and gog sheets get.\n
  • Boundary markers: No delimiters or specific instructions are implemented to isolate external data from the agent's core operating instructions.\n
  • Capability inventory: The agent possesses the capability to perform high-impact actions, such as sending emails (gog gmail send) or modifying spreadsheets, based on instructions that could be embedded in the retrieved documents or messages.\n
  • Sanitization: There is no evidence of data validation or filtering for malicious instructions within the content fetched from external services.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 18, 2026, 11:27 PM