The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it extracts data from potentially untrusted Excel files and presents it to the agent without sanitization or protective boundary markers.
Ingestion points: scripts/read_excel.py and scripts/analyze_excel.py read and return values from spreadsheet cells.
Boundary markers: Data is passed as raw strings within JSON responses without specific delimiters or instructions to ignore embedded commands.
Capability inventory: The skill has the capability to read local files via the openpyxl library in scripts/read_excel.py and scripts/analyze_excel.py.
Sanitization: No content validation or sanitization is performed on the cell values before they are returned to the agent context.

excel-processor