skills/szweibel/claude-skills/xlsx/Gen Agent Trust Hub

xlsx

Pass

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The recalc.py script executes the soffice binary through the subprocess module to automate spreadsheet recalculation tasks.
  • [REMOTE_CODE_EXECUTION]: The skill performs dynamic code generation by writing a StarBasic macro file (Module1.xba) to the user's local LibreOffice configuration directory and subsequently executing it.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its processing of untrusted external spreadsheet files. • Ingestion points: Files loaded using pandas.read_excel and openpyxl.load_workbook as described in SKILL.md. • Boundary markers: No markers or instructions are defined to separate untrusted data from the agent's instructions. • Capability inventory: Command execution via the soffice utility and file system write access for macro generation. • Sanitization: There is no evidence of content validation or formula sanitization for the spreadsheets being processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 20, 2026, 12:45 AM
Security Audit — agent-trust-hub — xlsx