The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes specialized MCP tools such as wps_get_active_workbook and wps_execute_method to perform extensive operations within the WPS Spreadsheet application. These operations include reading cell values, writing formulas, and deleting rows or sheets.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it retrieves and processes untrusted data from spreadsheet cells which could contain malicious instructions aimed at the agent.
Ingestion points: Data enters the agent's context from active workbooks through tools and methods like wps_get_active_workbook, wps_get_cell_value, and getRangeData as described in SKILL.md.
Boundary markers: The skill does not define any boundary markers or instructions to treat spreadsheet content as untrusted data rather than commands.
Capability inventory: The agent has significant permissions to modify, delete, and overwrite workbook data and structure via the wps_execute_method interface.
Sanitization: There are no sanitization or verification steps applied to data retrieved from spreadsheets before it is used to generate or execute solutions.

wps-excel