hydrogen-markets-localization

Pass

Audited by Gen Agent Trust Hub on Sep 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill directs the agent to execute local utility scripts (e.g., node scripts/search_shopify_docs.mjs and node scripts/search_weaverse_docs.mjs) to retrieve the latest platform documentation for Shopify and Weaverse.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data from URL segments, Shopify Admin responses, and CMS content. It implements a thorough safety strategy including boundary enforcement (rejecting malformed locale shapes), capability inventory (restricted GraphQL proxy), and sanitization (validating redirect targets to ensure they are app-relative and redacting sensitive data from logs).
  • [DATA_EXFILTRATION]: The skill utilizes a GraphQL API proxy on the vendor's official domain (studio.weaverse.io) to interact with Shopify Admin. This communication is consistent with the skill's purpose and the author's identity.
  • [CREDENTIALS_UNSAFE]: Includes explicit security instructions for handling the WEAVERSE_API_KEY, requiring its storage in environment files with restricted permissions (0600) and prohibiting its exposure in prompts, documentation, or logs.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 11, 2026, 03:16 AM
Security Audit — agent-trust-hub — hydrogen-markets-localization