excalidraw-skill

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill's Step 0 explicitly instructs cloning a public GitHub repo (https://github.com/wise-toddler/mcp_excalidraw) and its workflow requires calling describe_scene / get_canvas_screenshot to read arbitrary, user-editable canvas content which the agent is expected to interpret and act on, so public third-party content can materially influence tool use and decisions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill's auto-setup explicitly runs "git clone https://github.com/wise-toddler/mcp_excalidraw" followed by npm install/npm run build and executing launch.sh, which fetches and executes remote code at runtime (https://github.com/wise-toddler/mcp_excalidraw), creating a clear high-risk external dependency.