linkfox-sorftime-amazon-product-query
Pass
Audited by Gen Agent Trust Hub on Aug 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes Amazon product data retrieved from an external API (
tool-gateway.linkfox.com). This represents an ingestion surface for untrusted external content. - Ingestion points: Data is fetched via the
productsarray in thescripts/sorftime_product_search.pyscript. - Boundary markers: The
SKILL.mdprovides display rules instructing the agent to present data in clear tables and use specific extraction tools likejqto limit context exposure. - Capability inventory: The skill has the capability to perform network operations, write data files to the local project directory, and execute shell commands via Python scripts.
- Sanitization: The provided Python scripts perform basic JSON parsing and summarization but do not implement explicit sanitization or filtering of text fields (like titles or brand names) before they enter the agent's context.
- [PERSISTENCE_MECHANISMS]: The
references/onboarding.mdfile provides manual instructions for the user to modify shell initialization files (~/.bashrc,~/.zshrc, or Windows environment variables) to store theLINKFOX_AGENT_API_KEY. While this is a standard configuration practice for API keys, it involves modifying persistence files to maintain session data. - [COMMAND_EXECUTION]: The skill relies on two Python scripts (
scripts/sorftime_product_search.pyandscripts/onboarding.py) to perform its core functions. These scripts are executed by the agent to interact with external APIs, manage authentication tokens, and write query results to a session-specific data directory within the project folder.
Audit Metadata