presentation-creation

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions and examples are strictly focused on presentation design, narrative structure, and output formatting. No malicious code, network exfiltration, or unauthorized file access was detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided content (outlines and documents) to generate presentation outputs, which is a potential surface for indirect prompt injection. 1. Ingestion points: The skill accepts 'Source material' such as outlines or document paths as input for presentation generation. 2. Boundary markers: The provided prompt template lacks clear delimiters (e.g., XML tags or block markers) to isolate untrusted user data from the system's instructions. 3. Capability inventory: The skill is capable of generating structured markup (HTML/reveal.js) and code (python-pptx scripts). 4. Sanitization: There are no explicit instructions for the agent to sanitize the input data or to disregard instructions embedded within the source materials.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 09:01 PM
Security Audit — agent-trust-hub — presentation-creation