create-deck

Pass

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill utilizes local CLI tools (create-deck, add-slide, navigate) for deck management. These tools are vendor-provided and operate within the project context without unauthorized command execution.- [PROMPT_INJECTION]: The skill features an indirect prompt injection surface by processing external data from Google Docs via the import-google-doc tool (SKILL.md). This untrusted content is interpolated into slide templates via add-slide (SKILL.md) without specific boundary markers or sanitization steps described. This is a common pattern for integration-focused skills.- [SAFE]: The skill uses static HTML templates for slide layouts. There is no evidence of code obfuscation, runtime code compilation, or unsafe dynamic execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 13, 2026, 04:51 AM
Security Audit — agent-trust-hub — create-deck