commerce-integrated-strategy

Pass

Audited by Gen Agent Trust Hub on Jun 22, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it is designed to ingest and process untrusted external data.
  • Ingestion points: The skill explicitly requests and processes user-provided dashboard screenshots, CSV/Excel report exports from external commerce platforms (Naver, Coupang), and outputs from preceding workflow skills (SKILL.md).
  • Boundary markers: The instructions do not define explicit delimiters or 'ignore' instructions to isolate user-provided data from the agent's core instructions.
  • Capability inventory: The skill generates comprehensive strategy reports, creates JSON structures for macro strategies, and performs automatic chaining to the moai-core:ai-slop-reviewer skill (SKILL.md).
  • Sanitization: There is no evidence of content sanitization or validation for the data ingested from external sources.
  • [SAFE]: All external URL references in the documentation point to well-known Korean e-commerce management services such as ShopManager, EasySeller, and Sabangnet (references/channel-mix.md).
  • [SAFE]: The skill uses automatic chaining to vendor-specific tools (e.g., moai-core:ai-slop-reviewer) which is consistent with the provided author context for 'modu-ai' (SKILL.md).
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 22, 2026, 06:27 AM
Security Audit — agent-trust-hub — commerce-integrated-strategy