linkfox-shopee-product-detail
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONPERSISTENCE
Full Analysis
- [COMMAND_EXECUTION]: The skill uses Python scripts (
scripts/shopee_product_detail.pyandscripts/onboarding.py) to handle API interactions and user account management. - [EXTERNAL_DOWNLOADS]: The scripts communicate with LinkFox's official domains (
tool-gateway.linkfox.com,api.linkfox.com,agent-api.linkfox.com) to fetch product data and perform account operations. These are vendor-owned resources used for the skill's primary purpose. - [INDIRECT_PROMPT_INJECTION]: The skill processes Shopee product data which could contain untrusted instructions.
- Ingestion points: Product details (names, descriptions, attributes) are retrieved from Shopee URLs via a proxy API and ingested into the agent's context.
- Boundary markers: The instructions guide the agent to present structured data but do not implement specific delimiters or warnings for untrusted text content.
- Capability inventory: The skill has the capability to write JSON data files to a local 'linkfox' directory within the project.
- Sanitization: The scripts parse JSON responses but do not explicitly sanitize or filter string fields before they are presented to the agent.
- [PERSISTENCE]: The onboarding documentation contains instructions for users to manually add their API keys to shell profile files (
.bashrc,.zshrc) to ensure the keys are persisted for future sessions.
Audit Metadata