presentation-design

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill instructions do not contain any malicious patterns such as credential theft, privilege escalation, or persistence mechanisms. Tools are used according to their intended design-related purposes.
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface because it processes untrusted user-provided content to generate design plans.
  • Ingestion points: User inputs for the deck's topic, audience, and slide text are extracted and used to build the plan in SKILL.md.
  • Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands in the processed user data.
  • Capability inventory: The skill uses the image_gen tool and writes to the local file system (presentations/ directory).
  • Sanitization: No sanitization or validation of the input strings is performed before processing.
  • Mitigation: The risk is mitigated by a mandatory human-in-the-loop checkpoint (Step 5), where the user must review and approve the deck plan before the agent proceeds with image generation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 04:39 AM
Security Audit — agent-trust-hub — presentation-design