pp-breezedoc

Pass

Audited by Gen Agent Trust Hub on Jul 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions to download and install executable components from external sources, including npm and GitHub.\n
  • Evidence: npx -y @mvanhorn/printing-press-library install breezedoc --cli-only\n
  • Evidence: go install github.com/mvanhorn/printing-press-library/library/productivity/breezedoc/cmd/breezedoc-pp-cli@latest\n
  • Context: These resources are hosted under the author's namespace (mvanhorn).\n- [COMMAND_EXECUTION]: The skill facilitates the execution of shell commands and external binaries for management and installation.\n
  • Evidence: Direct execution of breezedoc-pp-cli for API interaction.\n
  • Evidence: Use of system tools like npx, go, and claude for setup tasks.\n- [DATA_EXFILTRATION]: The CLI tool supports a delivery mechanism that allows output to be routed to external webhooks.\n
  • Evidence: The --deliver webhook:<url> flag can be used to POST command results to arbitrary external endpoints.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because user-provided input is interpolated directly into command-line executions.\n
  • Ingestion points: User input is captured via the $ARGUMENTS variable defined in the skill documentation.\n
  • Boundary markers: None identified; user input is not delimited from command instructions.\n
  • Capability inventory: The skill has the capability to execute the breezedoc-pp-cli binary, which can access the Breezedoc API and deliver results to external files or webhooks.\n
  • Sanitization: No validation or sanitization steps are defined for the $ARGUMENTS input before it is passed to the shell.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 28, 2026, 09:18 AM
Security Audit — agent-trust-hub — pp-breezedoc