pivot-table-generator
Warn
Audited by Gen Agent Trust Hub on Feb 15, 2026
Risk Level: MEDIUMPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (MEDIUM): The skill processes untrusted external CSV data which can be used to inject instructions into the agent's context during data analysis.
- Ingestion points: The
scripts/pivot_table_generator.pyfile reads external data usingpd.read_csv(args.data)at line 86. - Boundary markers: Absent. The data is parsed directly into a pandas DataFrame without delimiters or instructions to the agent to ignore embedded content.
- Capability inventory: The script has the capability to read local files via
pd.read_csvand write to the local filesystem usingto_csv,to_excel,to_html, andplt.savefig(lines 53-61, 70-76). - Sanitization: Absent. The script relies on standard pandas parsing which does not filter for Natural Language instructions embedded in data fields.
- [Data Exposure] (LOW): The script accepts arbitrary file paths via CLI arguments for both input (
--data) and output (--output). An attacker could potentially trick the agent into reading sensitive configuration files as CSVs or overwriting critical files with pivot table data.
Audit Metadata