wps-office
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a comprehensive suite of 231 tools to control local WPS Office applications. Tools such as
wps_word_open_document,wps_excel_write_range, andwps_common_save_asallow the agent to perform extensive file system operations and application-level commands. While these are primary features of the skill, they represent a high-privilege capability set within the local environment. - [INDIRECT_PROMPT_INJECTION]:
- Ingestion points: Data enters the agent context through tools like
wps_word_get_document_text,wps_excel_read_range, andwps_ppt_get_slide_info, which read content directly from user-provided office files. - Boundary markers: The instructions lack specific boundary markers or warnings to the agent to treat document content as potentially untrusted data or to ignore embedded instructions found within those documents.
- Capability inventory: The skill maintains a large inventory of capabilities, including batch file conversion, automated document generation, and structural modifications to the file system (e.g.,
wps_convert_format,wps_common_save_as). - Sanitization: There is no evidence of sanitization or validation logic to prevent content read from a document (e.g., an Excel cell containing a malicious instruction) from influencing the agent's next action or subsequent tool parameters.
Audit Metadata