excalidraw

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
  • [NO_CODE] (SAFE): The skill consists entirely of markdown instructions and task templates without any scripts, executables, or binary data.
  • [Indirect Prompt Injection] (LOW): The skill involves processing user-provided .excalidraw files via subagents. 1. Ingestion points: .excalidraw and .excalidraw.json files are read by subagents. 2. Boundary markers: Absent from the subagent task templates. 3. Capability inventory: Read and Write tools are used to process and modify diagrams. 4. Sanitization: Absent; the templates focus on token extraction without mentions of content sanitization or instruction filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 04:47 PM