pp-klaviyo

Pass

Audited by Gen Agent Trust Hub on May 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Fetches and installs the klaviyo-pp-cli and klaviyo-pp-mcp binaries from the vendor's GitHub repository (github.com/mvanhorn/printing-press-library) using the go install command.
  • [EXTERNAL_DOWNLOADS]: Downloads the installation utility via npx from the @mvanhorn/printing-press package on NPM.
  • [DATA_EXFILTRATION]: The skill features an output delivery flag --deliver webhook:<url> which enables the routing of command results (potentially containing customer data or API responses) to arbitrary external URLs.
  • [COMMAND_EXECUTION]: The skill's core functionality is built around executing the klaviyo-pp-cli binary through the Bash tool based on user-provided arguments.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the way it processes untrusted data.
  • Ingestion points: Processes external files such as HTML templates (--template-html) and marketing briefs (--brief), in addition to data retrieved from the Klaviyo API.
  • Boundary markers: Absent. The instructions do not define delimiters or provide warnings to the agent regarding instructions that might be embedded in the ingested data.
  • Capability inventory: The skill utilizes the Bash tool, can write to the local file system (--deliver file:<path>), and perform network POST requests (--deliver webhook:<url>).
  • Sanitization: Absent. There are no specified procedures for validating or sanitizing the content of external files or API responses before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
May 8, 2026, 06:02 PM