excalidraw-learning-aids

Pass

Audited by Gen Agent Trust Hub on May 11, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: Detailed review of the skill's instructions and utility scripts confirms that they operate within expected parameters. No malicious patterns, such as prompt injection, persistence mechanisms, or credential theft, were identified.
  • [PROMPT_INJECTION]: The skill processes untrusted input from local repositories, creating an inherent surface for indirect prompt injection, although no specific malicious instructions are present in the provided files.
  • Ingestion points: Repository source code, manifests, and file structures are processed by scripts/repo_snapshot.py and direct agent inspection.
  • Boundary markers: The instructions do not specify delimiters or warnings to ignore instructions found within the analyzed source code.
  • Capability inventory: The agent can execute the provided Python scripts (repo_snapshot.py, create_excalidraw_scene.py) and interact with Excalidraw-related tools or APIs.
  • Sanitization: Content from the repository is read without sanitization or validation to filter out potential natural language instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
May 11, 2026, 07:46 AM