The Agent Skills Directory

[SAFE]: The skill relies on the well-known and trusted openpyxl library for all spreadsheet operations. Dependencies are managed via uv, a standard and secure installation method.
[PROMPT_INJECTION]: The skill possesses an attack surface for indirect prompt injection as it processes data from external files that could contain malicious instructions.
Ingestion points: Data is read from external .xlsx workbooks and JSON specification files in scripts/inspect_xlsx.py, scripts/edit_xlsx.py, and scripts/create_xlsx.py.
Boundary markers: The instructions do not define clear delimiters or use 'ignore embedded instructions' warnings when the agent processes contents from spreadsheet cells, which could allow malicious data to influence agent behavior.
Capability inventory: The skill provides file read and write capabilities via the openpyxl library. It does not possess network access or arbitrary command execution capabilities, limiting the potential impact of an injection.
Sanitization: The scripts perform JSON parsing and basic type coercion, but do not sanitize the text content of cells to prevent them from being interpreted as instructions by the agent.

xlsx