linkfox-shopify-product-query
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [DATA_EXPOSURE_AND_EXFILTRATION]: The skill's onboarding process, specifically within
scripts/onboarding.pyandreferences/onboarding.md, encourages the collection of the user's phone number to facilitate SMS-based registration and login for the LinkFox service. - [PERSISTENCE_MECHANISMS]: The
references/onboarding.mdfile contains instructions for the AI agent to guide the user in modifying their system's environment. It explicitly suggests appending configuration to shell profile files such as~/.zshrcand~/.bashrc, or usingsetxon Windows to ensure theLINKFOX_AGENT_API_KEYpersists across sessions. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted product data from Shopify and writes the full responses to the local file system in the
linkfox/directory. - Ingestion points: External data enters through
ehunt/shopify/productQueryAPI calls inscripts/shopify_product_query.py. - Boundary markers: No boundary markers or 'ignore' instructions are used when handling the product titles, descriptions, or store information.
- Capability inventory: The skill performs file writes to the local disk and is used in contexts where an agent might later read and process these files.
- Sanitization: There is no evidence of sanitization or filtering of the content returned from the external Shopify API before storage or display.
- [EXTERNAL_DOWNLOADS]: The
scripts/onboarding.pyscript requires external Python packages (requests,qrcode,pillow) and instructs the agent or user to install them if missing, which introduces third-party code into the execution environment.
Audit Metadata