skills/tt-a1i/hermes-agent/excalidraw/Gen Agent Trust Hub

excalidraw

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The scripts/upload.py utility transmits diagram data to excalidraw.com. This is the official sharing endpoint for a well-known diagramming service. The implementation follows security best practices by using AES-GCM encryption to protect the content, ensuring the encryption key remains in the URL fragment and is not readable by the server.
  • [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting data to generate diagrams. If the agent is instructed to visualize untrusted content (such as web pages or documents), that content could contain malicious instructions designed to influence the agent's behavior during the diagramming process.
  • Ingestion points: Data or text provided to the agent for visualization.
  • Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present in the diagram generation templates.
  • Capability inventory: The skill is capable of writing files to the local disk and executing the upload.py script via the terminal.
  • Sanitization: No explicit sanitization or filtering of input data is performed before it is structured into the Excalidraw JSON format.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 12:47 PM
Security Audit — agent-trust-hub — excalidraw