sales-data-extraction-agent
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security findings detected. The skill performs legitimate sales data extraction and persistence.
- [PROMPT_INJECTION]: The skill processes external Excel files, which is an attack surface for indirect prompt injection. However, no specific vulnerabilities or malicious instructions were found in the implementation logic. The risk is inherent to the skill's data processing function. * Ingestion points: Local filesystem watcher for Excel (.xlsx, .xls) files in the SKILL.md defined workflow. * Boundary markers: None explicitly defined in the prompt instructions to delimit data from instructions. * Capability inventory: PostgreSQL database insertion and local file system reading via the run.sh script and linked runner. * Sanitization: The skill employs fuzzy column mapping for data normalization, but does not specify content sanitization for instructions embedded within cells.
Audit Metadata