pptx

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes several local utility scripts (e.g., unpack.py, pack.py, rearrange.py, inventory.py, replace.py) and system tools (soffice, pdftoppm) for legitimate document management and conversion tasks.
  • [PROMPT_INJECTION]: The skill maintains a surface for indirect prompt injection due to its core function of ingesting and processing untrusted data from external PowerPoint files. This risk is addressed by recommending the use of defusedxml for secure XML parsing.
  • Ingestion points: PowerPoint files are processed using the markitdown tool and raw XML unpacking scripts as described in skill.md.
  • Boundary markers: The instructions do not specify explicit delimiters to isolate extracted text from the agent's instruction context.
  • Capability inventory: The skill facilitates extensive local file system access and the execution of document processing utilities.
  • Sanitization: The skill explicitly includes defusedxml as a dependency to protect against XML-based injection and expansion attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 10:17 AM
Security Audit — agent-trust-hub — pptx