skills/tjxj/z-skills/z-excel-editor/Gen Agent Trust Hub

z-excel-editor

Warn

Audited by Gen Agent Trust Hub on Jun 25, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the subprocess module to execute several external system commands. These include soffice (LibreOffice) for formula recalculation, git for comparing document text, and gcc for compiling a compatibility shim.
  • Evidence in scripts/recalc.py, scripts/office/soffice.py, and scripts/office/validators/redlining.py.
  • [REMOTE_CODE_EXECUTION]: To bypass restrictions on Unix sockets in certain sandboxed environments, the skill dynamically generates C source code at runtime, compiles it into a shared object library using gcc, and uses the LD_PRELOAD environment variable to inject this shim into the LibreOffice process.
  • Evidence in scripts/office/soffice.py.
  • [PROMPT_INJECTION]: No patterns were detected that attempt to override agent instructions or bypass safety filters.
  • [DATA_EXFILTRATION]: No evidence of hardcoded credentials or unauthorized network operations was found. The skill uses defusedxml for XML parsing, which is a security best practice for preventing XML External Entity (XXE) attacks when handling Office documents.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 25, 2026, 02:22 PM
Security Audit — agent-trust-hub — z-excel-editor