docx
Warn
Audited by Gen Agent Trust Hub on Apr 1, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs runtime compilation of C source code. In
scripts/office/soffice.py, the script writes a socket shim (lo_socket_shim.c) to a temporary directory and executesgccto compile it into a shared library. This library is then dynamically loaded viaLD_PRELOADto modify system call behavior for LibreOffice. - [COMMAND_EXECUTION]: The skill generates and executes application macros. In
scripts/accept_changes.py, it writes a StarBasic macro to the filesystem and directs thesofficebinary to execute it, which allows for automated document manipulation. - [COMMAND_EXECUTION]: The skill extensively utilizes the
subprocessmodule to execute system binaries includingsoffice,pandoc,pdftoppm,git, andgccto perform core document management and environment setup tasks. - [EXTERNAL_DOWNLOADS]: The skill instructions guide the user to install external dependencies such as the
docxJavaScript library from the global npm registry. - [PROMPT_INJECTION]: The skill has a high surface area for indirect prompt injection. It ingests and processes untrusted data from
.docxfiles without implementing boundary markers or specific instructions to the agent to ignore potentially malicious embedded commands within the document XML.
Audit Metadata