pp-coffee-goat

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install its core CLI and MCP server components from the author's GitHub and NPM namespaces.
  • Evidence: Installation via npx -y @mvanhorn/printing-press-library and go install github.com/mvanhorn/printing-press-library/....
  • Analysis: These are recognized as vendor resources associated with the author 'mvanhorn' as specified in the provided context.
  • [COMMAND_EXECUTION]: The skill operates by executing the coffee-goat-pp-cli binary to process user requests.
  • Evidence: Direct Use instructions and numerous command examples such as coffee-goat god-cup --method espresso --agent.
  • Analysis: The agent is instructed to execute shell commands with user-provided arguments, utilizing an --agent flag for structured output.
  • [DATA_EXFILTRATION]: The CLI includes a built-in feature to deliver command output directly to external webhooks.
  • Evidence: The --deliver webhook:<url> flag documentation in SKILL.md.
  • Analysis: This is a documented feature for data routing but provides a functional mechanism for sending processed data to arbitrary network endpoints.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 05:30 PM
Security Audit — agent-trust-hub — pp-coffee-goat