ai-presentation-maker

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a centralized helper script (assets/presentation-helper.sh) to perform file system and system-level operations. This script implements internal sanitization and path validation to prevent command injection and directory traversal.\n- [EXTERNAL_DOWNLOADS]: Generated HTML slides reference external assets from well-known and trusted services, including Google Fonts and Cloudflare (FontAwesome). These are standard dependencies for the visual presentation layer.\n- [PROMPT_INJECTION]: The system instructions include strict operational rules (e.g., 'Use EXACT text', 'NEVER tell the user to open a terminal') aimed at ensuring consistent behavior and preventing the agent from deviating from its intended function. No malicious bypass attempts were detected.\n- [DATA_EXFILTRATION]: All data storage and processing are restricted to the local ~/workspace/presentations/ directory. The skill requests access to 'AI Persona OS' files (SOUL.md, AGENTS.md) for profile integration, which is a documented feature and remains within the local workspace environment.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 10:49 AM
Security Audit — agent-trust-hub — ai-presentation-maker