shopify-redirect-mapping
Pass
Audited by Gen Agent Trust Hub on Jul 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides detailed logic and algorithms, such as fuzzy matching implementations and GraphQL mutation recipes, for the agent to implement as executable code during a session to manage store data.\n- [EXTERNAL_DOWNLOADS]: The instructions recommend installing the Shopify-AI-Toolkit from the official Shopify GitHub organization to facilitate Admin API interactions.\n- [PROMPT_INJECTION]: The skill defines specific behavioral constraints and logic instructions, such as the "recall over precision" doctrine and explicit rules against homepage fallbacks, to guide the agent's reasoning.\n- [PROMPT_INJECTION]: The skill processes external CSV files containing URLs, presenting an indirect prompt injection surface.\n
- Ingestion points: references/pipeline.md (Step 1) reads user-provided 404 exports.\n
- Boundary markers: The skill relies on normalization and classification (Step 2) to prepare data, though no specific delimiters or "ignore instructions" warnings are mentioned for the URL content.\n
- Capability inventory: The agent has write access to the Shopify store's navigation via GraphQL mutations (urlRedirectCreate, urlRedirectUpdate) in references/pipeline.md (Step 6 & 7).\n
- Sanitization: Standard URL normalization (lowercasing, stripping protocol, and dropping query strings) is performed in Step 2 before processing.\n- [SAFE]: The skill includes explicit security warnings against hardcoding sensitive Shopify Admin API tokens, providing clear instructions for secure management via environment variables.
Audit Metadata