pp-marianatek

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the marianatek-pp-cli tool. It provides instructions to fetch this from the official @mvanhorn/printing-press npm package or the vendor's GitHub repository at github.com/mvanhorn/printing-press-library.- [COMMAND_EXECUTION]: The skill executes shell commands using the marianatek-pp-cli binary to search for classes, manage reservations, and check account status across multiple studio tenants.- [DATA_EXFILTRATION]: The integrated CLI tool includes a --deliver webhook:<url> feature. While this is a functional capability designed for automation and routing command results, it allows for the transmission of reservation and account data to external endpoints.- [PROMPT_INJECTION]: The skill ingests class schedules, instructor names, and studio metadata from the Mariana Tek API, which constitutes a potential surface for indirect prompt injection.
  • Ingestion points: Data retrieved from the Mariana Tek API via marianatek-pp-cli search and schedule commands.
  • Boundary markers: No explicit instruction-ignoring delimiters are mentioned for the external data, though the output is processed as structured JSON.
  • Capability inventory: The skill uses the Bash tool and performs network operations through the CLI binary.
  • Sanitization: No explicit sanitization of API-returned strings is described in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 05:19 PM
Security Audit — agent-trust-hub — pp-marianatek