shoplazza-shop

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were identified. The skill's operations are aligned with its purpose as a store management tool.
  • [COMMAND_EXECUTION]: The skill uses the shoplazza CLI to interact with store data. It includes safety practices like using --dry-run and requiring user confirmation for destructive operations.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes data from the store (e.g., metafields, analytics) which could contain malicious instructions. 1. Ingestion points: Commands that read store data such as shop info get, shop analytics, and shop metafields list in SKILL.md and reference files. 2. Boundary markers: Not explicitly documented in these reference files. 3. Capability inventory: CLI commands for creating, updating, and deleting store resources as described in all files. 4. Sanitization: The skill relies on the CLI and the agent's safety protocols.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 01:10 AM
Security Audit — agent-trust-hub — shoplazza-shop