skills/notabanker/codexclawd/gog/Gen Agent Trust Hub

gog

Pass

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is built around executing the gog command-line utility to perform all interactions with Google Workspace services.- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the gog tool via Homebrew from a third-party repository (steipete/tap/gogcli).- [DATA_EXFILTRATION]: The skill has extensive access to sensitive user data, including the ability to search and read Gmail messages, list contacts, search Google Drive, and export Documents. It also possesses the capability to send emails and modify spreadsheet data, which are core functions of the tool.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes content from external sources (Gmail, Sheets, Drive) that may contain adversarial instructions.
  • Ingestion points: Content is ingested via Gmail search, spreadsheet reading, and document export commands in SKILL.md.
  • Boundary markers: No specific delimiters or safety instructions are provided to the agent to differentiate between tool output and instructions.
  • Capability inventory: The skill allows for the execution of arbitrary gog commands, which can modify data or send communications.
  • Sanitization: External content is processed without explicit sanitization or validation of the retrieved text strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 10, 2026, 01:26 AM
Security Audit — agent-trust-hub — gog