linkfox-sif-keyword-summary
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes product information (such as titles and descriptions) from Amazon via LinkFox APIs. This data is untrusted and provides a vector for indirect prompt injection.\n
- Ingestion points: Data returned from the
/sif/keywordSummaryAPI endpoint, processed inscripts/sif_keyword_traffic.py.\n - Boundary markers: There are no defined delimiters to isolate untrusted external content from the agent's instructions.\n
- Capability inventory: The skill has capabilities for file system writes (data storage) and network access to vendor-owned APIs.\n
- Sanitization: No evidence of filtering or escaping external metadata was found in the scripts.\n- [COMMAND_EXECUTION]: The skill utilizes local Python scripts (
scripts/sif_keyword_traffic.pyandscripts/onboarding.py) to execute its core logic and manage user authentication. These scripts are invoked via shell commands by the agent platform.\n- [EXTERNAL_DOWNLOADS]: The skill interacts with various vendor-owned domains (e.g.,tool-gateway.linkfox.com,api.linkfox.com,agent-api.linkfox.com) and references external Python dependencies (requests,qrcode,pillow) required for its onboarding and payment functionality.
Audit Metadata