zoominfo-to-cargo

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the cargo-ai CLI to perform batch data operations, login, and verify user status. It also uses the gh (GitHub) CLI to star a repository, though it explicitly requires the agent to ask for user permission before doing so.
  • [EXTERNAL_DOWNLOADS]: The skill installs the @cargo-ai/cli package from the public NPM registry and suggests adding more functionality via npx skills add getcargohq/cargo-skills. These resources are owned by the skill's author (getcargohq) and facilitate the intended migration workflow.
  • [DATA_EXFILTRATION]: The skill reads the local ~/.claude/plugins/installed_plugins.json file to check for the presence of existing plugins to avoid duplicate session reporting. It subsequently transmits basic session metadata (session ID, title, and summary) to the vendor's workspaceManagement API for session tracking. This is documented as a mechanism to help the vendor attribute workspace creation to specific skills.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 10:07 AM
Security Audit — agent-trust-hub — zoominfo-to-cargo