skills/jd-opensource/joysafeter/pptx/Gen Agent Trust Hub

pptx

Fail

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: HIGHCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
  • [Command Execution] (HIGH): The ooxml/scripts/unpack.py script uses zipfile.ZipFile.extractall() without path validation. This is a Zip Slip vulnerability that allows a malicious Office document to overwrite arbitrary files on the system via path traversal (e.g., using '..' in filenames).
  • [Data Exfiltration] (MEDIUM): The ooxml/scripts/validation/docx.py module uses lxml.etree.parse() with default settings, which is susceptible to XML External Entity (XXE) attacks. This could allow an attacker to read local files or perform SSRF if the agent processes a specially crafted document.
  • [Indirect Prompt Injection] (LOW): The skill processes untrusted external document data which could contain malicious instructions designed to influence the agent's behavior. Evidence: 1. Ingestion points: ooxml/scripts/unpack.py and various validation scripts processing .docx/.pptx files; 2. Boundary markers: None observed; 3. Capability inventory: File reading/writing via zip extraction and subprocess execution of soffice; 4. Sanitization: Inconsistent use of defusedxml (used in packing but not in docx validation).
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 18, 2026, 05:57 PM