The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from Word documents, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: The skill invokes wps_word_get_document_text and wps_word_get_paragraphs (SKILL.md) to retrieve content from existing documents into the agent's context.
Boundary markers: There are no explicit instructions or delimiters defined to help the agent distinguish between user instructions and potentially malicious data embedded within the documents.
Capability inventory: The skill has significant capabilities to modify files, open new documents via wps_word_open_document, and insert external assets via wps_word_insert_image (SKILL.md).
Sanitization: The instructions do not specify any validation or sanitization logic for the data retrieved from documents before it is used to drive subsequent tool calls.
[COMMAND_EXECUTION]: The skill enables automated execution of a wide range of Word document commands and file system operations (wps_word_open_document, wps_word_insert_image) based on natural language processing, which increases the impact if the agent is successfully manipulated.

wps-word