skills/anbeime/skill/ppt-generator/Gen Agent Trust Hub

ppt-generator

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied themes and role-generated content to create structured data, representing an attack surface for indirect prompt injection.\n
  • Ingestion points: The skill ingests user input and agent-generated text to populate ppt_data.json for script processing.\n
  • Boundary markers: Detailed JSON schema definitions are provided in references/ppt_structure_guide.md, and the generation script implements field and layout type validation.\n
  • Capability inventory: The scripts/generate_pptx.py script has the capability to perform filesystem writes to create .pptx files.\n
  • Sanitization: Structural validation is present in the code, though specific textual sanitization for prompt injection patterns is not explicitly implemented.\n- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute a local Python script (scripts/generate_pptx.py) to build the final presentation, which involves interaction with the local filesystem and the use of the python-pptx library.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 10:52 AM