skills/agentic-os-org/anolisa/xlsx/Gen Agent Trust Hub

xlsx

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the subprocess module in its helper scripts to perform automated spreadsheet tasks. Specifically, scripts/xlsx_insert_row.py executes xlsx_shift_rows.py to reindex cell references, and scripts/libreoffice_recalc.py executes the soffice binary for headless formula evaluation. These subprocess calls use argument lists rather than shell strings, which follows security best practices to prevent command injection.
  • [INDIRECT_PROMPT_INJECTION]: As a tool designed to read and process external data files, the skill naturally has an attack surface for indirect prompt injection.
  • Ingestion points: Data enters the agent's context through xlsx_reader.py (which uses pandas) and through the unpacking of XLSX archives in xlsx_unpack.py.
  • Boundary markers: The skill instructions do not specify explicit delimiters or "ignore" instructions for the untrusted data it processes.
  • Capability inventory: The skill utilizes shell execution (via companion scripts) and file writing to generate and modify XLSX files.
  • Sanitization: scripts/xlsx_unpack.py includes a robust path validation check that prevents zip-slip attacks by ensuring extracted files cannot be written outside the designated temporary working directory.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 03:58 AM