skills/sharadchaturveda-coder/agency-agents-codex/agency-sales-data-extraction-agent/Gen Agent Trust Hub
agency-sales-data-extraction-agent
Pass
Audited by Gen Agent Trust Hub on Mar 25, 2026
Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
- [PROMPT_INJECTION]: The skill's primary function is to monitor and parse external Excel files (.xlsx, .xls), which introduces an indirect prompt injection surface. 1. Ingestion points: Local filesystem monitoring of designated directories for new sales reports, as defined in SKILL.md. 2. Boundary markers: The instructions do not specify any delimiters or explicit commands to ignore instructions that might be embedded within the Excel data. 3. Capability inventory: The skill utilizes filesystem monitoring and read operations, along with PostgreSQL database write access for bulk inserts and transactions, as detailed in SKILL.md. 4. Sanitization: While the skill mentions validating extracted metrics, it lacks specific measures to sanitize or filter potential natural language instructions embedded within cell contents, sheet names, or metadata.
- [NO_CODE]: This skill consists solely of markdown instructions and configuration files; it does not include any executable scripts or source code.
Audit Metadata