linkfox-shopify-store-query
Warn
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: MEDIUMPERSISTENCEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PERSISTENCE]: The skill directs the agent to modify user shell configurations (such as
.bashrcand.zshrc) and Windows environment settings (usingsetx) to store theLINKFOX_AGENT_API_KEYacross sessions. Additionally, theonboarding.pyscript maintains local state by writing session metadata and activity logs to a persistentlinkfox/directory in the workspace or home folder. - [INDIRECT_PROMPT_INJECTION]: The skill processes store metadata (e.g., store names, domains) fetched from the Shopify tool gateway, which creates a potential injection surface.
- Ingestion points: Data enters the agent's context through the JSON response of
scripts/shopify_store_query.py. - Boundary markers: The skill does not employ explicit delimiters or instructions to ignore embedded prompts in the Shopify data.
- Capability inventory: The skill has capabilities for network requests, file system writes, and shell command execution.
- Sanitization: There is no evidence of sanitization for the external store data before it is presented to the agent.
- [COMMAND_EXECUTION]: The skill's scripts and documentation involve the execution of shell commands to manage environment variables and interact with remote API services for store queries and user onboarding.
- [EXTERNAL_DOWNLOADS]: The skill references the installation of standard Python packages, specifically
requests,qrcode, andpillow, from well-known public registries to support its login and payment QR code generation features.
Audit Metadata