pp-elevenlabs

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

• Third-party content exposure detected (high risk: 0.90). The skill's CLI explicitly includes commands that scrape and ingest public webpages/URLs into agent-accessible knowledge (e.g., "convai create-url-document-route" and "convai refresh-url-document-route", plus "audio-native update-content-from-url" and "studio add-project" which can initialize from a URL), so the agent will read untrusted, user-provided web content that can affect subsequent tool behavior.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill requires a local CLI that the docs instruct to install at runtime by fetching and executing remote packages (e.g., via npx -y @mvanhorn/printing-press install elevenlabs and go install github.com/mvanhorn/printing-press-library/library/ai/elevenlabs/cmd/elevenlabs-pp-cli@latest and github.com/mvanhorn/printing-press-library/library/ai/elevenlabs/cmd/elevenlabs-pp-mcp@latest), which pulls and runs remote code that the skill depends on.